Mapping and Analyzing Geospatial Trends: A Python Approach

By

Imonikhe Ayeni

---

INTRODUCTION

This project, titled "Mapping and Analyzing Geospatial Trends: A Python Approach" explores the integration of geospatial data visualization and sentiment analysis using Python. The work is organized into three distinct tasks, each focusing on a specific aspect of geospatial data handling:

• Geospatial Visualization:

This task demonstrates the use of Python-based geospatial visualization tools, specifically GeoPandas, to analyze and present real-world datasets. GeoPandas facilitates the creation of both static and interactive maps, providing a comprehensive view of the geospatial information.

• Geospatial Data Analysis:

This task delves into professional methods for analyzing geospatial datasets. It utilizes the World Total Population dataset from the World Bank and the World Cereal Yield dataset. The objective is to identify significant relationships between the population of different countries and their cereal yield, showcasing the potential correlations and insights that can be derived from such analyses.

• Geospatial Sentiment Analysis:

In this task, geospatial sentiment analysis is applied to Twitter (now X) data using the Python library, TextBlob. The dataset comprises tweets relevant to cryptocurrency. This task aims to extract and map sentiments expressed in these tweets, highlighting the geographical distribution of opinions and trends in the cryptocurrency domain.

OBJECTIVES

The objectives of this project are as follows:

• Identify the concepts Underlying Geospatial Analysis:

Identify the concepts underlying geospatial analysis through the application of Python-based tools and libraries.

• Apply Social Analytics Techniques:

Implement appropriate techniques to analyze social information, particularly through the sentiment analysis of Twitter data related to cryptocurrency.

• Design and Implement a Geospatial Analysis Framework:

Develop, prototype, and execute a comprehensive framework for geospatial analysis, integrating various datasets and visualization methods to derive meaningful insights.

• Identify Relationships Between Geospatial Datasets:

Investigate and identify significant relationships between different geospatial datasets, such as the correlation between the population of different countries and their cereal yield.

• Enhance Data Visualization Skills:

Improve skills in creating both static and interactive geospatial visualizations, making complex data more accessible and understandable to a diverse audience.

METHODOLOGY

The methodology for this project involves the following steps, corresponding to each of the three main tasks:

1. Geospatial Visualization

• Data Collection:

Obtain a real-world dataset suitable for geospatial analysis.

• Data Preparation:

Clean and preprocess the dataset to ensure it is ready for analysis using GeoPandas.

• Visualization:

Use GeoPandas' .plot() and .explore() methods to create both static and interactive maps. Employ additional libraries such as Folium and Mapclassify to enhance the interactivity and classification of the maps.

• Presentation:

Display the static maps for quick insights and interactive maps for detailed exploration, allowing users to hover, zoom, and interact with the geospatial data.

2. Geospatial Data Analysis

• Data Collection:

Source the World Total Population dataset from the World Bank and the World Cereal Yield dataset.

• Data Preparation:

Clean and preprocess the datasets to ensure compatibility for analysis.

• Analysis:

Use statistical and geospatial analysis techniques to explore the relationship between the population of different countries and their cereal yield. Employ visualization tools to present the findings in a clear and interpretable manner.

• Interpretation:

Analyze the results to identify significant correlations or trends, providing insights into the potential relationship between population and cereal yield.

3. Geospatial Sentiment Analysis

• Data Collection:

Collect tweets relevant to cryptocurrency from Twitter (now X).

• Data Preparation:

Clean and preprocess the tweet dataset, including text normalization and location extraction.

• Sentiment Analysis:

Use the TextBlob library to perform sentiment analysis on the tweets, categorizing them into positive, negative, and neutral sentiments.

• Geospatial Mapping:

Map the sentiments geographically using GeoPandas to visualize the distribution of opinions and trends across different regions.

• Interpretation:

Analyze the mapped sentiments to identify regional trends and insights related to cryptocurrency discussions on Twitter.

My first step will be to install my libraries. Notice the installation of folium and mapclassify they will help with interactive maps.

!pip install geopandas folium matplotlib mapclassify

Collecting geopandas
  Downloading geopandas-1.0.1-py3-none-any.whl.metadata (2.2 kB)
Collecting folium
  Downloading folium-0.17.0-py2.py3-none-any.whl.metadata (3.8 kB)
Requirement already satisfied: matplotlib in c:\users\user\anaconda3\lib\site-packages (3.8.4)
Collecting mapclassify
  Downloading mapclassify-2.8.0-py3-none-any.whl.metadata (2.8 kB)
Requirement already satisfied: numpy>=1.22 in c:\users\user\anaconda3\lib\site-packages (from geopandas) (1.26.4)
Collecting pyogrio>=0.7.2 (from geopandas)
  Downloading pyogrio-0.9.0-cp312-cp312-win_amd64.whl.metadata (3.9 kB)
Requirement already satisfied: packaging in c:\users\user\anaconda3\lib\site-packages (from geopandas) (23.2)
Requirement already satisfied: pandas>=1.4.0 in c:\users\user\anaconda3\lib\site-packages (from geopandas) (2.2.2)
Collecting pyproj>=3.3.0 (from geopandas)
  Downloading pyproj-3.6.1-cp312-cp312-win_amd64.whl.metadata (31 kB)
Collecting shapely>=2.0.0 (from geopandas)
  Downloading shapely-2.0.5-cp312-cp312-win_amd64.whl.metadata (7.2 kB)
Collecting branca>=0.6.0 (from folium)
  Downloading branca-0.7.2-py3-none-any.whl.metadata (1.5 kB)
Requirement already satisfied: jinja2>=2.9 in c:\users\user\anaconda3\lib\site-packages (from folium) (3.1.4)
Requirement already satisfied: requests in c:\users\user\anaconda3\lib\site-packages (from folium) (2.32.2)
Requirement already satisfied: xyzservices in c:\users\user\anaconda3\lib\site-packages (from folium) (2022.9.0)
Requirement already satisfied: contourpy>=1.0.1 in c:\users\user\anaconda3\lib\site-packages (from matplotlib) (1.2.0)
Requirement already satisfied: cycler>=0.10 in c:\users\user\anaconda3\lib\site-packages (from matplotlib) (0.11.0)
Requirement already satisfied: fonttools>=4.22.0 in c:\users\user\anaconda3\lib\site-packages (from matplotlib) (4.51.0)
Requirement already satisfied: kiwisolver>=1.3.1 in c:\users\user\anaconda3\lib\site-packages (from matplotlib) (1.4.4)
Requirement already satisfied: pillow>=8 in c:\users\user\anaconda3\lib\site-packages (from matplotlib) (10.3.0)
Requirement already satisfied: pyparsing>=2.3.1 in c:\users\user\anaconda3\lib\site-packages (from matplotlib) (3.0.9)
Requirement already satisfied: python-dateutil>=2.7 in c:\users\user\anaconda3\lib\site-packages (from matplotlib) (2.9.0.post0)
Requirement already satisfied: networkx>=2.7 in c:\users\user\anaconda3\lib\site-packages (from mapclassify) (3.2.1)
Requirement already satisfied: scikit-learn>=1.0 in c:\users\user\anaconda3\lib\site-packages (from mapclassify) (1.4.2)
Requirement already satisfied: scipy>=1.8 in c:\users\user\anaconda3\lib\site-packages (from mapclassify) (1.13.1)
Requirement already satisfied: MarkupSafe>=2.0 in c:\users\user\anaconda3\lib\site-packages (from jinja2>=2.9->folium) (2.1.3)
Requirement already satisfied: pytz>=2020.1 in c:\users\user\anaconda3\lib\site-packages (from pandas>=1.4.0->geopandas) (2024.1)
Requirement already satisfied: tzdata>=2022.7 in c:\users\user\anaconda3\lib\site-packages (from pandas>=1.4.0->geopandas) (2023.3)
Requirement already satisfied: certifi in c:\users\user\anaconda3\lib\site-packages (from pyogrio>=0.7.2->geopandas) (2024.7.4)
Requirement already satisfied: six>=1.5 in c:\users\user\anaconda3\lib\site-packages (from python-dateutil>=2.7->matplotlib) (1.16.0)
Requirement already satisfied: joblib>=1.2.0 in c:\users\user\anaconda3\lib\site-packages (from scikit-learn>=1.0->mapclassify) (1.4.2)
Requirement already satisfied: threadpoolctl>=2.0.0 in c:\users\user\anaconda3\lib\site-packages (from scikit-learn>=1.0->mapclassify) (2.2.0)
Requirement already satisfied: charset-normalizer<4,>=2 in c:\users\user\anaconda3\lib\site-packages (from requests->folium) (2.0.4)
Requirement already satisfied: idna<4,>=2.5 in c:\users\user\anaconda3\lib\site-packages (from requests->folium) (3.7)
Requirement already satisfied: urllib3<3,>=1.21.1 in c:\users\user\anaconda3\lib\site-packages (from requests->folium) (2.2.2)
Downloading geopandas-1.0.1-py3-none-any.whl (323 kB)
   ---------------------------------------- 0.0/323.6 kB ? eta -:--:--
   ---------------------------------------- 323.6/323.6 kB 9.8 MB/s eta 0:00:00
Downloading folium-0.17.0-py2.py3-none-any.whl (108 kB)
   ---------------------------------------- 0.0/108.4 kB ? eta -:--:--
   ---------------------------------------- 108.4/108.4 kB 6.1 MB/s eta 0:00:00
Downloading mapclassify-2.8.0-py3-none-any.whl (58 kB)
   ---------------------------------------- 0.0/58.9 kB ? eta -:--:--
   ---------------------------------------- 58.9/58.9 kB 3.0 MB/s eta 0:00:00
Downloading branca-0.7.2-py3-none-any.whl (25 kB)
Downloading pyogrio-0.9.0-cp312-cp312-win_amd64.whl (15.9 MB)
   ---------------------------------------- 0.0/15.9 MB ? eta -:--:--
   --- ------------------------------------ 1.2/15.9 MB 25.7 MB/s eta 0:00:01
   ------ --------------------------------- 2.6/15.9 MB 28.3 MB/s eta 0:00:01
   ---------- ----------------------------- 4.3/15.9 MB 30.7 MB/s eta 0:00:01
   -------------- ------------------------- 5.6/15.9 MB 29.9 MB/s eta 0:00:01
   ----------------- ---------------------- 6.9/15.9 MB 29.5 MB/s eta 0:00:01
   -------------------- ------------------- 8.2/15.9 MB 29.0 MB/s eta 0:00:01
   ----------------------- ---------------- 9.5/15.9 MB 28.8 MB/s eta 0:00:01
   -------------------------- ------------- 10.6/15.9 MB 29.7 MB/s eta 0:00:01
   ------------------------------ --------- 12.0/15.9 MB 28.4 MB/s eta 0:00:01
   -------------------------------- ------- 13.0/15.9 MB 28.4 MB/s eta 0:00:01
   ------------------------------------ --- 14.3/15.9 MB 26.2 MB/s eta 0:00:01
   ---------------------------------------  15.9/15.9 MB 27.3 MB/s eta 0:00:01
   ---------------------------------------  15.9/15.9 MB 27.3 MB/s eta 0:00:01
   ---------------------------------------  15.9/15.9 MB 27.3 MB/s eta 0:00:01
   ---------------------------------------- 15.9/15.9 MB 21.1 MB/s eta 0:00:00
Downloading pyproj-3.6.1-cp312-cp312-win_amd64.whl (6.1 MB)
   ---------------------------------------- 0.0/6.1 MB ? eta -:--:--
   --------- ------------------------------ 1.5/6.1 MB 47.2 MB/s eta 0:00:01
   ----------------- ---------------------- 2.7/6.1 MB 34.9 MB/s eta 0:00:01
   --------------------------- ------------ 4.1/6.1 MB 33.0 MB/s eta 0:00:01
   ---------------------------------- ----- 5.3/6.1 MB 30.7 MB/s eta 0:00:01
   ---------------------------------------  6.1/6.1 MB 32.4 MB/s eta 0:00:01
   ---------------------------------------- 6.1/6.1 MB 24.3 MB/s eta 0:00:00
Downloading shapely-2.0.5-cp312-cp312-win_amd64.whl (1.4 MB)
   ---------------------------------------- 0.0/1.4 MB ? eta -:--:--
   ----------------------------------- ---- 1.3/1.4 MB 40.6 MB/s eta 0:00:01
   ---------------------------------------- 1.4/1.4 MB 23.1 MB/s eta 0:00:00
Installing collected packages: shapely, pyproj, pyogrio, branca, mapclassify, geopandas, folium
Successfully installed branca-0.7.2 folium-0.17.0 geopandas-1.0.1 mapclassify-2.8.0 pyogrio-0.9.0 pyproj-3.6.1 shapely-2.0.5

I am going to be using interactive map in this work, thus I am importing folium and mapclassify. Pandas will help me setup and manage a dataframe. Geopandas will give me the geometry column this is necessary for mapping activities. Mapplotlib will assist me in visualisation. GeoPandas is a Python library essential for working with vector data. It is built on the pandas library. Crickard, P., (2018)

import geopandas  as gpd    # importing necassary dependancies
import pandas as pd
import matplotlib.pyplot as plt
import folium
import mapclassify

Task 1.1: Application of Python-based geospatial visualisation tool (e.g., GeoPandas) on a realworld dataset

INSTRUCTION

This task requires you to use the dataset, cereal yield. Use a Python-based visualisation tool (such as GeoPandas) to plot a set of choropleth maps representing the world cereal yield (kg per hectare) for the years 2019 and 2020 respectively. The solution should be in a Jupyter notebook (.ipynb), wherein all the functions, libraries and coding steps should be explained in a lucid manner. Major steps for generating the choropleths would typically involve, importing the datasets using appropriate Python libraries, data cleaning, geospatial operations, and plotting. The Jupyter Notebook should be able to reproduce the choropleth maps without any error.

df_cereal_yield = pd.read_csv(r'API_AG.YLD.CREL.KG_DS2_en_csv_v2_5734359.csv', skiprows=4) # Reading my data set with pandas library. I skipped some rows as they had unreadable or no value

df_cereal_yield.tail() #inspect dataset

	Country Name	Country Code	Indicator Name	Indicator Code	1960	1961	1962	1963	1964	1965	...	2014	2015	2016	2017	2018	2019	2020	2021	2022	Unnamed: 67
261	Kosovo	XKX	Cereal yield (kg per hectare)	AG.YLD.CREL.KG	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
262	Yemen, Rep.	YEM	Cereal yield (kg per hectare)	AG.YLD.CREL.KG	NaN	782.5	780.7	771.8	776.1	773.6	...	962.7	784.2	687.0	699.0	682.8	864.9	861.1	791.8	NaN	NaN
263	South Africa	ZAF	Cereal yield (kg per hectare)	AG.YLD.CREL.KG	NaN	1099.1	1142.1	1128.0	913.9	911.4	...	4899.6	3348.4	3623.1	5331.8	4652.1	4101.4	5120.6	5124.7	NaN	NaN
264	Zambia	ZMB	Cereal yield (kg per hectare)	AG.YLD.CREL.KG	NaN	822.2	801.4	706.9	788.9	823.5	...	2774.9	3026.4	2432.2	2489.9	2168.1	2400.4	2481.6	2525.0	NaN	NaN
265	Zimbabwe	ZWE	Cereal yield (kg per hectare)	AG.YLD.CREL.KG	NaN	919.7	905.9	822.5	820.5	930.8	...	831.4	557.5	435.1	1203.3	1254.3	748.0	1148.6	1545.2	NaN	NaN

5 rows × 68 columns

#Inspect Dataset

df_cereal_yield.columns

Index(['Country Name', 'Country Code', 'Indicator Name', 'Indicator Code',
       '1960', '1961', '1962', '1963', '1964', '1965', '1966', '1967', '1968',
       '1969', '1970', '1971', '1972', '1973', '1974', '1975', '1976', '1977',
       '1978', '1979', '1980', '1981', '1982', '1983', '1984', '1985', '1986',
       '1987', '1988', '1989', '1990', '1991', '1992', '1993', '1994', '1995',
       '1996', '1997', '1998', '1999', '2000', '2001', '2002', '2003', '2004',
       '2005', '2006', '2007', '2008', '2009', '2010', '2011', '2012', '2013',
       '2014', '2015', '2016', '2017', '2018', '2019', '2020', '2021', '2022',
       'Unnamed: 67'],
      dtype='object')

df_cereal_yield.isna().sum()

Country Name        0
Country Code        0
Indicator Name      0
Indicator Code      0
1960              266
                 ... 
2019               39
2020               39
2021               39
2022              266
Unnamed: 67       266
Length: 68, dtype: int64

#extract useful columns accoring to the instruction
useful_columns = ['Country Name','Country Code','2019', '2020']

cereal_yield = df_cereal_yield[useful_columns]

cereal_yield

	Country Name	Country Code	2019	2020
0	Aruba	ABW	NaN	NaN
1	Africa Eastern and Southern	AFE	1717.894885	1838.762607
2	Afghanistan	AFG	2113.400000	1979.900000
3	Africa Western and Central	AFW	1343.462790	1381.643141
4	Angola	AGO	958.800000	992.500000
...	...	...	...	...
261	Kosovo	XKX	NaN	NaN
262	Yemen, Rep.	YEM	864.900000	861.100000
263	South Africa	ZAF	4101.400000	5120.600000
264	Zambia	ZMB	2400.400000	2481.600000
265	Zimbabwe	ZWE	748.000000	1148.600000

266 rows × 4 columns

cereal_yield.isna().sum() #view null values

Country Name     0
Country Code     0
2019            39
2020            39
dtype: int64

I am going to drop columns with null values. I cannot fill the null values with mean or mode, because each country has its perculiarities.

cereal_yield.dropna(subset =['2019', '2020'], inplace= True)

C:\Users\User\AppData\Local\Temp\ipykernel_8832\632718052.py:1: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  cereal_yield.dropna(subset =['2019', '2020'], inplace= True)

cereal_yield

	Country Name	Country Code	2019	2020
1	Africa Eastern and Southern	AFE	1717.894885	1838.762607
2	Afghanistan	AFG	2113.400000	1979.900000
3	Africa Western and Central	AFW	1343.462790	1381.643141
4	Angola	AGO	958.800000	992.500000
5	Albania	ALB	5038.200000	5209.200000
...	...	...	...	...
259	World	WLD	4125.447810	4116.427597
262	Yemen, Rep.	YEM	864.900000	861.100000
263	South Africa	ZAF	4101.400000	5120.600000
264	Zambia	ZMB	2400.400000	2481.600000
265	Zimbabwe	ZWE	748.000000	1148.600000

227 rows × 4 columns

cereal_yield.isna().sum() #all null values removed

Country Name    0
Country Code    0
2019            0
2020            0
dtype: int64

I will now bring in geopandas dataset (naturaleath_lowres) and merge it with cereal yield

#check available geopandas datasets
gpd.datasets.available 

['naturalearth_cities', 'naturalearth_lowres', 'nybb']

#use naturalearth lowres
earth= gpd.read_file(gpd.datasets.get_path('naturalearth_lowres'))

C:\Users\User\AppData\Local\Temp\ipykernel_8832\3099934314.py:2: FutureWarning: The geopandas.dataset module is deprecated and will be removed in GeoPandas 1.0. You can get the original 'naturalearth_lowres' data from https://www.naturalearthdata.com/downloads/110m-cultural-vectors/.
  earth= gpd.read_file(gpd.datasets.get_path('naturalearth_lowres'))

earth

	pop_est	continent	name	iso_a3	gdp_md_est	geometry
0	889953.0	Oceania	Fiji	FJI	5496	MULTIPOLYGON (((180.00000 -16.06713, 180.00000...
1	58005463.0	Africa	Tanzania	TZA	63177	POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...
2	603253.0	Africa	W. Sahara	ESH	907	POLYGON ((-8.66559 27.65643, -8.66512 27.58948...
3	37589262.0	North America	Canada	CAN	1736425	MULTIPOLYGON (((-122.84000 49.00000, -122.9742...
4	328239523.0	North America	United States of America	USA	21433226	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...
...	...	...	...	...	...	...
172	6944975.0	Europe	Serbia	SRB	51475	POLYGON ((18.82982 45.90887, 18.82984 45.90888...
173	622137.0	Europe	Montenegro	MNE	5542	POLYGON ((20.07070 42.58863, 19.80161 42.50009...
174	1794248.0	Europe	Kosovo	-99	7926	POLYGON ((20.59025 41.85541, 20.52295 42.21787...
175	1394973.0	North America	Trinidad and Tobago	TTO	24269	POLYGON ((-61.68000 10.76000, -61.10500 10.890...
176	11062113.0	Africa	S. Sudan	SSD	11998	POLYGON ((30.83385 3.50917, 29.95350 4.17370, ...

177 rows × 6 columns

#filter out useful columns
earth= earth[["iso_a3","geometry"]]
earth

	iso_a3	geometry
0	FJI	MULTIPOLYGON (((180.00000 -16.06713, 180.00000...
1	TZA	POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...
2	ESH	POLYGON ((-8.66559 27.65643, -8.66512 27.58948...
3	CAN	MULTIPOLYGON (((-122.84000 49.00000, -122.9742...
4	USA	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...
...	...	...
172	SRB	POLYGON ((18.82982 45.90887, 18.82984 45.90888...
173	MNE	POLYGON ((20.07070 42.58863, 19.80161 42.50009...
174	-99	POLYGON ((20.59025 41.85541, 20.52295 42.21787...
175	TTO	POLYGON ((-61.68000 10.76000, -61.10500 10.890...
176	SSD	POLYGON ((30.83385 3.50917, 29.95350 4.17370, ...

177 rows × 2 columns

#rename "iso_a3" to Country Code to create a common column with cereal_yield dataframe 
earth= earth.rename(columns={"iso_a3":'Country Code'})

earth #Inspect

	Country Code	geometry
0	FJI	MULTIPOLYGON (((180.00000 -16.06713, 180.00000...
1	TZA	POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...
2	ESH	POLYGON ((-8.66559 27.65643, -8.66512 27.58948...
3	CAN	MULTIPOLYGON (((-122.84000 49.00000, -122.9742...
4	USA	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...
...	...	...
172	SRB	POLYGON ((18.82982 45.90887, 18.82984 45.90888...
173	MNE	POLYGON ((20.07070 42.58863, 19.80161 42.50009...
174	-99	POLYGON ((20.59025 41.85541, 20.52295 42.21787...
175	TTO	POLYGON ((-61.68000 10.76000, -61.10500 10.890...
176	SSD	POLYGON ((30.83385 3.50917, 29.95350 4.17370, ...

177 rows × 2 columns

#we shall merge earth data frame and cereal_yield dataframe and call the eventual dataframe earth_yield
earth_yield= earth.merge(cereal_yield, on='Country Code')

earth_yield

	Country Code	geometry	Country Name	2019	2020
0	FJI	MULTIPOLYGON (((180.00000 -16.06713, 180.00000...	Fiji	3353.4	3665.7
1	TZA	POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...	Tanzania	1883.3	1698.4
2	CAN	MULTIPOLYGON (((-122.84000 49.00000, -122.9742...	Canada	4010.8	4095.4
3	USA	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...	United States	8006.1	8145.3
4	KAZ	POLYGON ((87.35997 49.21498, 86.59878 48.54918...	Kazakhstan	1154.5	1288.6
...	...	...	...	...	...
162	MKD	POLYGON ((22.38053 42.32026, 22.88137 41.99930...	North Macedonia	3554.0	3664.2
163	SRB	POLYGON ((18.82982 45.90887, 18.82984 45.90888...	Serbia	6126.9	6559.4
164	MNE	POLYGON ((20.07070 42.58863, 19.80161 42.50009...	Montenegro	3166.8	3260.0
165	TTO	POLYGON ((-61.68000 10.76000, -61.10500 10.890...	Trinidad and Tobago	1529.7	1520.3
166	SSD	POLYGON ((30.83385 3.50917, 29.95350 4.17370, ...	South Sudan	880.5	885.5

167 rows × 5 columns

I will sort my dataframe in descending order to show top five countries with highest cereal yield and bottom five countries with lowest cereal yield

earth_yield.sort_values(by='2019', ascending= False)

	Country Code	geometry	Country Name	2019	2020
79	ARE	POLYGON ((51.57952 24.24550, 51.75744 24.29407...	United Arab Emirates	23842.3	25980.3
81	KWT	POLYGON ((47.97452 29.97582, 48.18319 29.53448...	Kuwait	17625.4	13393.1
83	OMN	MULTIPOLYGON (((55.20834 22.70833, 55.23449 23...	Oman	13291.0	18835.1
124	BEL	POLYGON ((6.15666 50.80372, 6.04307 50.12805, ...	Belgium	8988.5	8430.6
125	NLD	POLYGON ((6.90514 53.48216, 7.09205 53.14404, ...	Netherlands	8654.2	7919.9
...	...	...	...	...	...
84	VUT	MULTIPOLYGON (((167.21680 -15.89185, 167.84488...	Vanuatu	618.2	609.7
13	SDN	POLYGON ((24.56737 8.22919, 23.80581 8.66632, ...	Sudan	552.9	489.6
11	SOM	POLYGON ((41.58513 -1.68325, 40.99300 -0.85829...	Somalia	522.2	502.5
51	NER	POLYGON ((14.85130 22.86295, 15.09689 21.30852...	Niger	501.5	560.2
75	GMB	POLYGON ((-16.71373 13.59496, -15.62460 13.623...	Gambia, The	443.8	501.9

167 rows × 5 columns

earth_yield.sort_values(by='2020', ascending= False)

	Country Code	geometry	Country Name	2019	2020
79	ARE	POLYGON ((51.57952 24.24550, 51.75744 24.29407...	United Arab Emirates	23842.3	25980.3
83	OMN	MULTIPOLYGON (((55.20834 22.70833, 55.23449 23...	Oman	13291.0	18835.1
81	KWT	POLYGON ((47.97452 29.97582, 48.18319 29.53448...	Kuwait	17625.4	13393.1
131	NZL	MULTIPOLYGON (((176.88582 -40.06598, 176.50802...	New Zealand	8205.0	9039.3
129	NCL	POLYGON ((165.77999 -21.08000, 166.59999 -21.7...	New Caledonia	7031.1	8689.9
...	...	...	...	...	...
11	SOM	POLYGON ((41.58513 -1.68325, 40.99300 -0.85829...	Somalia	522.2	502.5
75	GMB	POLYGON ((-16.71373 13.59496, -15.62460 13.623...	Gambia, The	443.8	501.9
13	SDN	POLYGON ((24.56737 8.22919, 23.80581 8.66632, ...	Sudan	552.9	489.6
22	LSO	POLYGON ((28.97826 -28.95560, 29.32517 -29.257...	Lesotho	695.5	433.3
46	NAM	POLYGON ((19.89577 -24.76779, 19.89473 -28.461...	Namibia	731.5	429.5

167 rows × 5 columns

Fig 1: World cereal yield (kg per hectare) for the year 2019 with .plot

fig, ax = plt.subplots(figsize=(20,16))
earth.plot(ax=ax, color= "black")
earth_yield.plot(ax=ax, column="2019", legend=True, legend_kwds={"label": "World cereal yield (kg per hectare) for the year 2019","orientation":"horizontal"}, cmap='Set1')
plt.show()

fig, ax = plt.subplots(figsize=(20,16))
earth.plot(ax=ax, color= "black")
earth_yield.plot(ax=ax, column="2020", legend=True, legend_kwds={"label": "World cereal yield (kg per hectare) for the year 2020","orientation":"horizontal"}, cmap='Set1')
plt.show()

I am now going to use .explore for my visualisation. The motive behind using this interactive map in explaned on methodoloy and the libraries to install to use .explore are also highlighted, with reference(s) provided.

earth_yield.explore(column='2019',  # make choropleth based on "2019" column
    tooltip=["Country Name",'2019'],  # show "country name and 2019 value" in tooltip (on hover)
    popup=True,  # show all values in popup (on click)
    tiles="openstreetmap",  # I shall use "openstreetmap" tiles
    cmap="Set1",  # I shall use "Set1" matplotlib colormap
    legend=True,                
    style_kwds=dict(color="black"), # use black outline
    zoom_control=True,
                   )                

Make this Notebook Trusted to load map: File -> Trust Notebook

earth_yield.explore(column='2019',  # make choropleth based on "2019" column
    tooltip=["Country Name",'2019'],  # show "country name and 2019 value" in tooltip (on hover)
    popup=True,  # show all values in popup (on click)
    tiles="",  # I shall use "CartoDB positron" tiles
    cmap="Set1",  # I shall use "Set1" matplotlib colormap
    legend=True,                
    style_kwds=dict(color="black"), # use black outline
    zoom_control=True,
                   )    

Make this Notebook Trusted to load map: File -> Trust Notebook

Brief Explanation:

From the tables and maps, it is obvious that:

1. The top three countries with the highest cereal yield are in Asia for 2019 and 2020.
2. The United Arab Emirates tops the list for both 2019 and 2020 with 23842.3 and 25980.3kg per hectare, respectively.
3. It will be observed that the UAE grew its yield between 2019 and 2020.
4. The 4th and 5th countries are in Europe: Belgium at 8988.5kg per hectare and the Netherlands at 8654.2kg per hectare, respectively, for 2019.
5. Belgium and the Netherlands dropped in cereal yield between 2019 and 2020 and lost their position to New Zealand and New Caledonia.
6. Policymakers may want to investigate why they dropped and devise measures to make cereal yields sustainable in Belgium and the Netherlands.
7. African countries are sitting at the bottom for both 2019 and 2020.**

Task 1.2: Analysis of geospatial datasets

In this task, you are required to use one more dataset, the world's total population (source: World Bank) in addition to the cereal yield dataset used in the previous task. Both datasets are available on Moodle under the Assessment folder. All the choropleths and plots must be generated using appropriate Python-based tool

df_total_pop = pd.read_csv(r"API_SP.POP.TOTL_DS2_en_csv_v2_4485025.csv", skiprows=4) # Read population dataframe

df_total_pop[:3] #Inspect dataframe

	Country Name	Country Code	Indicator Name	Indicator Code	1960	1961	1962	1963	1964	1965	...	2013	2014	2015	2016	2017	2018	2019	2020	2021	Unnamed: 66
0	Aruba	ABW	Population, total	SP.POP.TOTL	54208.0	55434.0	56234.0	56699.0	57029.0	57357.0	...	103165.0	103776.0	104339.0	104865.0	105361.0	105846.0	106310.0	106766.0	107195.0	NaN
1	Africa Eastern and Southern	AFE	Population, total	SP.POP.TOTL	130836765.0	134159786.0	137614644.0	141202036.0	144920186.0	148769974.0	...	562601578.0	578075373.0	593871847.0	609978946.0	626392880.0	643090131.0	660046272.0	677243299.0	694665117.0	NaN
2	Afghanistan	AFG	Population, total	SP.POP.TOTL	8996967.0	9169406.0	9351442.0	9543200.0	9744772.0	9956318.0	...	32269592.0	33370804.0	34413603.0	35383028.0	36296111.0	37171922.0	38041757.0	38928341.0	39835428.0	NaN

3 rows × 67 columns

df_total_pop.columns

Index(['Country Name', 'Country Code', 'Indicator Name', 'Indicator Code',
       '1960', '1961', '1962', '1963', '1964', '1965', '1966', '1967', '1968',
       '1969', '1970', '1971', '1972', '1973', '1974', '1975', '1976', '1977',
       '1978', '1979', '1980', '1981', '1982', '1983', '1984', '1985', '1986',
       '1987', '1988', '1989', '1990', '1991', '1992', '1993', '1994', '1995',
       '1996', '1997', '1998', '1999', '2000', '2001', '2002', '2003', '2004',
       '2005', '2006', '2007', '2008', '2009', '2010', '2011', '2012', '2013',
       '2014', '2015', '2016', '2017', '2018', '2019', '2020', '2021',
       'Unnamed: 66'],
      dtype='object')

My next step will be to filter out all columns that will be useful to answer all questions for this task

necessary_columns = ['Country Name','Country Code', '2010', '2011', '2012', '2013',
       '2014', '2015', '2016', '2017', '2018', '2019', '2020', '2021'] 

total_pop=df_total_pop[necessary_columns]

total_pop.isna().sum()#I shall ignore the null values in this case as they have no direct impact on my work

Country Name    0
Country Code    0
2010            1
2011            1
2012            2
2013            2
2014            2
2015            2
2016            2
2017            2
2018            2
2019            2
2020            2
2021            2
dtype: int64

total_pop[:2] #inspect dataframe

	Country Name	Country Code	2010	2011	2012	2013	2014	2015	2016	2017	2018	2019	2020	2021
0	Aruba	ABW	101665.0	102050.0	102565.0	103165.0	103776.0	104339.0	104865.0	105361.0	105846.0	106310.0	106766.0	107195.0
1	Africa Eastern and Southern	AFE	518468229.0	532760424.0	547482863.0	562601578.0	578075373.0	593871847.0	609978946.0	626392880.0	643090131.0	660046272.0	677243299.0	694665117.0

total_pop.columns #confirming I have picked all useful columns

Index(['Country Name', 'Country Code', '2010', '2011', '2012', '2013', '2014',
       '2015', '2016', '2017', '2018', '2019', '2020', '2021'],
      dtype='object')

cereal_yield2= df_cereal_yield[necessary_columns] # I have recall my df_cereal_yield dataframe from Task 1, I will extract the necessary columns and strore in a fresh data frame

cereal_yield2[:2]

	Country Name	Country Code	2010	2011	2012	2013	2014	2015	2016	2017	2018	2019	2020	2021
0	Aruba	ABW	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
1	Africa Eastern and Southern	AFE	1553.204407	1492.710513	1650.256751	1533.952962	1636.00973	1616.362162	1490.807738	1764.116707	1728.295922	1717.894885	1838.762607	1840.899744

My next step will be to merge the two dataframe together, I have given them suffxies to avoid confusion

pop_yield = cereal_yield2.merge(total_pop, on='Country Code', suffixes=(['_yield','_pop']))

pop_yield[:3]

	Country Name_yield	Country Code	2010_yield	2011_yield	2012_yield	2013_yield	2014_yield	2015_yield	2016_yield	2017_yield	...	2012_pop	2013_pop	2014_pop	2015_pop	2016_pop	2017_pop	2018_pop	2019_pop	2020_pop	2021_pop
0	Aruba	ABW	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	102565.0	103165.0	103776.0	104339.0	104865.0	105361.0	105846.0	106310.0	106766.0	107195.0
1	Africa Eastern and Southern	AFE	1553.204407	1492.710513	1650.256751	1533.952962	1636.00973	1616.362162	1490.807738	1764.116707	...	547482863.0	562601578.0	578075373.0	593871847.0	609978946.0	626392880.0	643090131.0	660046272.0	677243299.0	694665117.0
2	Afghanistan	AFG	2011.100000	1659.900000	2029.600000	2048.500000	2017.50000	2132.200000	1980.400000	2022.500000	...	31161378.0	32269592.0	33370804.0	34413603.0	35383028.0	36296111.0	37171922.0	38041757.0	38928341.0	39835428.0

3 rows × 27 columns

Next step is to recall my earth dataframe and merge with pop_yield dataframe (containing population and cereal yield). Merger will be on thier common country code column

earth_yield2= earth.merge(pop_yield, on='Country Code', 
)

earth_yield2

	Country Code	geometry	Country Name_yield	2010_yield	2011_yield	2012_yield	2013_yield	2014_yield	2015_yield	2016_yield	...	2012_pop	2013_pop	2014_pop	2015_pop	2016_pop	2017_pop	2018_pop	2019_pop	2020_pop	2021_pop
0	FJI	MULTIPOLYGON (((180.00000 -16.06713, 180.00000...	Fiji	2871.1	2821.3	2423.4	2999.5	4475.4	3001.0	3000.9	...	865065.0	865602.0	866447.0	868632.0	872406.0	877460.0	883490.0	889955.0	896444.0	902899.0
1	TZA	POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...	Tanzania	1647.9	1390.4	1314.8	1418.0	1524.8	1449.4	1570.4	...	47053033.0	48483132.0	49960563.0	51482638.0	53049231.0	54660345.0	56313444.0	58005461.0	59734213.0	61498438.0
2	CAN	MULTIPOLYGON (((-122.84000 49.00000, -122.9742...	Canada	3501.1	3552.4	3456.3	4160.4	3647.0	3673.0	4239.1	...	34714222.0	35082954.0	35437435.0	35702908.0	36109487.0	36545236.0	37065084.0	37601230.0	38037204.0	38246108.0
3	USA	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...	United States	6978.1	6803.5	5911.9	7300.9	7638.1	7430.1	8614.2	...	313877662.0	316059947.0	318386329.0	320738994.0	323071755.0	325122128.0	326838199.0	328329953.0	331501080.0	331893745.0
4	KAZ	POLYGON ((87.35997 49.21498, 86.59878 48.54918...	Kazakhstan	804.1	1688.6	865.0	1164.9	1172.7	1278.1	1347.7	...	16792090.0	17035551.0	17288285.0	17542806.0	17794055.0	18037776.0	18276452.0	18513673.0	18755666.0	19002586.0
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
164	MKD	POLYGON ((22.38053 42.32026, 22.88137 41.99930...	North Macedonia	3329.6	3502.1	2839.3	3381.3	3900.0	3051.1	3859.2	...	2061044.0	2064032.0	2067471.0	2070226.0	2072490.0	2074502.0	2076217.0	2076694.0	2072531.0	2065092.0
165	SRB	POLYGON ((18.82982 45.90887, 18.82984 45.90888...	Serbia	4958.8	4751.4	3701.7	5157.8	5960.6	4787.3	6165.9	...	7199077.0	7164132.0	7130576.0	7095383.0	7058322.0	7020858.0	6982604.0	6945235.0	6899126.0	6844078.0
166	MNE	POLYGON ((20.07070 42.58863, 19.80161 42.50009...	Montenegro	3321.4	3305.7	2638.8	3770.5	3451.5	3146.5	3261.7	...	620601.0	621207.0	621810.0	622159.0	622303.0	622373.0	622227.0	622028.0	621306.0	620173.0
167	TTO	POLYGON ((-61.68000 10.76000, -61.10500 10.890...	Trinidad and Tobago	1667.4	1639.5	1471.7	1610.5	1329.4	1110.4	1444.3	...	1344814.0	1353708.0	1362337.0	1370332.0	1377563.0	1384060.0	1389841.0	1394969.0	1399491.0	1403374.0
168	SSD	POLYGON ((30.83385 3.50917, 29.95350 4.17370, ...	South Sudan	NaN	NaN	705.2	765.6	1253.7	907.3	879.1	...	10113648.0	10355030.0	10554882.0	10715657.0	10832520.0	10910774.0	10975924.0	11062114.0	11193729.0	11381377.0

169 rows × 28 columns

Task 1.2.1

For the year 2021, generate choropleth maps of cereal yield for only the countries having a population less than or equal to 67326569. Very briefly interpret the generated map

I have my dataframe well figured out, I shall now begin to respond to the questions. I shall filter according to question

earth_yield2.columns

Index(['Country Code', 'geometry', 'Country Name_yield', '2010_yield',
       '2011_yield', '2012_yield', '2013_yield', '2014_yield', '2015_yield',
       '2016_yield', '2017_yield', '2018_yield', '2019_yield', '2020_yield',
       '2021_yield', 'Country Name_pop', '2010_pop', '2011_pop', '2012_pop',
       '2013_pop', '2014_pop', '2015_pop', '2016_pop', '2017_pop', '2018_pop',
       '2019_pop', '2020_pop', '2021_pop'],
      dtype='object')

# I will drop all null values at this point
earth_yield2.dropna(inplace= True)

#filter for 2021
earth_yield_2021 = earth_yield2[['Country Code', 'geometry', 'Country Name_yield', '2021_yield', '2021_pop']]

earth_yield_2021 # This dataframe has all countries their population and yield for 2021. I shall filter according to task from here

	Country Code	geometry	Country Name_yield	2021_yield	2021_pop
0	FJI	MULTIPOLYGON (((180.00000 -16.06713, 180.00000...	Fiji	3887.3	902899.0
1	TZA	POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...	Tanzania	1651.1	61498438.0
2	CAN	MULTIPOLYGON (((-122.84000 49.00000, -122.9742...	Canada	3078.3	38246108.0
3	USA	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...	United States	8268.0	331893745.0
4	KAZ	POLYGON ((87.35997 49.21498, 86.59878 48.54918...	Kazakhstan	1049.0	19002586.0
...	...	...	...	...	...
163	BIH	POLYGON ((18.56000 42.65000, 17.67492 43.02856...	Bosnia and Herzegovina	4422.1	3263459.0
164	MKD	POLYGON ((22.38053 42.32026, 22.88137 41.99930...	North Macedonia	3537.8	2065092.0
165	SRB	POLYGON ((18.82982 45.90887, 18.82984 45.90888...	Serbia	5768.6	6844078.0
166	MNE	POLYGON ((20.07070 42.58863, 19.80161 42.50009...	Montenegro	3330.7	620173.0
167	TTO	POLYGON ((-61.68000 10.76000, -61.10500 10.890...	Trinidad and Tobago	1588.1	1403374.0

164 rows × 5 columns

#filter for population less than or equal to 67326569
pop_67326569 =earth_yield_2021[earth_yield_2021['2021_pop']<=67326569]

I am going to sort by 2021 yield and 2021 population, this is to show top and bottom countries and to assist me in explaining.

pop_67326569.sort_values(by='2021_yield', ascending= False) #filter by 2021 cereal yield

	Country Code	geometry	Country Name_yield	2021_yield	2021_pop
81	ARE	POLYGON ((51.57952 24.24550, 51.75744 24.29407...	United Arab Emirates	26226.2	9991083.0
85	OMN	MULTIPOLYGON (((55.20834 22.70833, 55.23449 23...	Oman	16461.4	5223376.0
83	KWT	POLYGON ((47.97452 29.97582, 48.18319 29.53448...	Kuwait	11216.7	4328553.0
133	NZL	MULTIPOLYGON (((176.88582 -40.06598, 176.50802...	New Zealand	8728.4	5122600.0
130	IRL	POLYGON ((-6.19788 53.86757, -6.03299 53.15316...	Ireland	8606.5	5028230.0
...	...	...	...	...	...
13	SDN	POLYGON ((24.56737 8.22919, 23.80581 8.66632, ...	Sudan	566.8	44909351.0
47	NAM	POLYGON ((19.89577 -24.76779, 19.89473 -28.461...	Namibia	517.3	2587344.0
11	SOM	POLYGON ((41.58513 -1.68325, 40.99300 -0.85829...	Somalia	502.6	16359500.0
77	GMB	POLYGON ((-16.71373 13.59496, -15.62460 13.623...	Gambia, The	490.7	2486937.0
52	NER	POLYGON ((14.85130 22.86295, 15.09689 21.30852...	Niger	349.6	25130810.0

143 rows × 5 columns

pop_67326569.sort_values(by='2021_pop', ascending= False) #sort descending by 2021 population

	Country Code	geometry	Country Name_yield	2021_yield	2021_pop
139	GBR	MULTIPOLYGON (((-6.19788 53.86757, -6.95373 54...	United Kingdom	6966.8	67326569.0
1	TZA	POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...	Tanzania	1651.1	61498438.0
22	ZAF	POLYGON ((16.34498 -28.57671, 16.82402 -28.082...	South Africa	5124.7	60041996.0
137	ITA	MULTIPOLYGON (((10.44270 46.89355, 11.04856 46...	Italy	5562.8	59066225.0
12	KEN	POLYGON ((39.20222 -4.67677, 37.76690 -3.67712...	Kenya	1487.6	54985702.0
...	...	...	...	...	...
145	BRN	POLYGON ((115.45071 5.44773, 115.40570 4.95523...	Brunei Darussalam	2885.2	441532.0
36	BLZ	POLYGON ((-89.14308 17.80832, -89.15091 17.955...	Belize	4289.5	404915.0
18	BHS	MULTIPOLYGON (((-78.98000 26.79000, -78.51000 ...	Bahamas, The	8419.2	396914.0
86	VUT	MULTIPOLYGON (((167.21680 -15.89185, 167.84488...	Vanuatu	609.5	314464.0
131	NCL	POLYGON ((165.77999 -21.08000, 166.59999 -21.7...	New Caledonia	7810.9	272620.0

143 rows × 5 columns

fig, ax = plt.subplots(figsize=(20,16))
earth.plot(ax=ax, color= "black")
pop_67326569.plot(ax=ax, column="2021_yield", legend=True, legend_kwds={"label": "cereal yield for only the countries having a population less than or equal to 67326569","orientation":"horizontal"}, cmap='Set1')
plt.show()

pop_67326569.explore(column='2021_yield',   # make choropleth based on "" column
    tooltip=["Country Name_yield",'2021_yield','2021_pop'],  # show "Country Name_yield",'2021_yield','2021_pop' values in tooltip (on hover)
    popup=True,  # show all values in popup (on click)
    tiles="openstreetmap",  # I shall use "openstreetmap" tiles
    cmap="Set1",  # use "Set1" matplotlib colormap
    legend=True,                
    style_kwds=dict(color="black"), # use black outline
    zoom_control=True,
    legend_kwds={"label": "cereal yield for only the countries having a population less than or equal to 67326569","orientation":"horizontal"}
                    )
    
                    

Make this Notebook Trusted to load map: File -> Trust Notebook

Brief Explanation

1. The United Arab Emirates at 26226.2kg per hectare, Oman at 16461.4kg per hectare, and Kuwait, all Asian countries, come first, second, and third for countries with the highest cereal yield for the year 2021.
2. The top 3 countries are in Asia.
3. The Australian continent is represented among this cohort by New Zealand at 8728.4kg per hectare.
4. Europe is represented in this cohort by the Republic of Ireland, at 8606.5kg per hectare.
5. Africa is at the bottom.

Policy implications and lessons learned.

The UAE, Oman, and Kuwait are striving to achieve self-sufficiency in food production on a sustainable scale. Great Britain, despite having the highest population, is not among the top 5. Great Britain should begin to study the models of the top countries in cereal production and begin to implement them immediately.

Given the rate of political unrest all over the world with allies being formed and the uncertainty of the global market, reliance on the importation of cereal and holding to the economic principles of specialization and international trade-offs may spell doom when push comes to shove.

Self-sufficiency in food production for Great Britain and all countries is a clarion call.

Task 1.2.2

For the year 2021, generate choropleth maps of cereal yield for only the countries having a population greater than or equal to 331,893,745. Very briefly interpret the generated map.

#filter for population greater than or equal to greater than or equal to 331893745
pop_331893745 = earth_yield_2021[earth_yield_2021['2021_pop']>=331893745]

pop_331893745

	Country Code	geometry	Country Name_yield	2021_yield	2021_pop
3	USA	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...	United States	8268.0	3.318937e+08
95	IND	POLYGON ((97.32711 28.26158, 97.40256 27.88254...	India	3478.8	1.393409e+09
136	CHN	MULTIPOLYGON (((109.47521 18.19770, 108.65521 ...	China	6320.8	1.412360e+09

fig, ax = plt.subplots(figsize=(20,16))
earth.plot(ax=ax, color= "black")
pop_331893745.plot(ax=ax, column="2021_yield", legend=True, legend_kwds={"label": "Creal yield for the countries having a population greater than or equal to 331,893,745","orientation":"horizontal"}, cmap='Set1')
plt.show()

pop_331893745.explore(column='2021_yield',
    tooltip=["Country Name_yield",'2021_yield','2021_pop'],  # show Country Name_yield",'2021_yield','2021_pop values in tooltip (on hover)
    popup=True,  # show all values in popup (on click)
    tiles="openstreetmap",  # use "openstreetmap" tiles
    cmap="Set1",  # use "Set1" matplotlib colormap
    legend=True,                
    style_kwds=dict(color="black"), # use black outline
    zoom_control=True,
    #zoom_start = 1,
   #scrollWheelZoom=False
                     )

Make this Notebook Trusted to load map: File -> Trust Notebook

Brief Explanation

The map shows that:

The United States, in gray, has a population of about 331 million people and a cereal yield of 8268. This cereal is just fair given its population.

China, in yellow, has a cereal yield of 6320.8kg per hectare and a population of 1.412 billion people as of 2021.

I have a bit of concern for India. I do not think that a cereal yield of 3478.8kg is right for a population of 1.393 billion. All hands have to be on deck to increase cereal yield to avoid Robert Thomas Malthus's 'doom' theory of a geometric population growth rate and arithmetic growth in food supplies (Intelligent Economist, 2020).

I am aware that there could be controversy over population control measures. But we can never go wrong with increasing cereal yield.

Task 1.2.3

For the year 2021, generate choropleth maps of cereal yield for only the countries having a population between 10269022 and 1393409034. Very briefly interpret the generated map

pop_10269022 = earth_yield_2021[(earth_yield_2021["2021_pop"]>= 10269022) & (earth_yield_2021["2021_pop"] <=1393409034)]

pop_10269022.sort_values(by='2021_yield', ascending= False) #sort descending by 2021 cereal yield

	Country Code	geometry	Country Name_yield	2021_yield	2021_pop
3	USA	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...	United States	8268.0	331893745.0
126	BEL	POLYGON ((6.15666 50.80372, 6.04307 50.12805, ...	Belgium	7906.2	11587882.0
127	NLD	POLYGON ((6.90514 53.48216, 7.09205 53.14404, ...	Netherlands	7872.3	17533405.0
40	FRA	MULTIPOLYGON (((-51.65780 4.15623, -52.24934 3...	France	7170.9	67499343.0
157	EGY	POLYGON ((36.86623 22.00000, 32.90000 22.00000...	Egypt, Arab Rep.	7132.5	104258327.0
...	...	...	...	...	...
14	TCD	POLYGON ((23.83766 19.58047, 23.88689 15.61084...	Chad	814.3	16914985.0
153	YEM	POLYGON ((52.00001 19.00000, 52.78218 17.34974...	Yemen, Rep.	791.8	30490639.0
13	SDN	POLYGON ((24.56737 8.22919, 23.80581 8.66632, ...	Sudan	566.8	44909351.0
11	SOM	POLYGON ((41.58513 -1.68325, 40.99300 -0.85829...	Somalia	502.6	16359500.0
52	NER	POLYGON ((14.85130 22.86295, 15.09689 21.30852...	Niger	349.6	25130810.0

87 rows × 5 columns

I will also sort by 2021 population to aid my discussion

pop_10269022.sort_values(by='2021_pop', ascending= False) #sort descending by 2021 population

	Country Code	geometry	Country Name_yield	2021_yield	2021_pop
95	IND	POLYGON ((97.32711 28.26158, 97.40256 27.88254...	India	3478.8	1.393409e+09
3	USA	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...	United States	8268.0	3.318937e+08
7	IDN	MULTIPOLYGON (((141.00021 -2.60015, 141.01706 ...	Indonesia	5351.3	2.763618e+08
99	PAK	POLYGON ((77.83745 35.49401, 76.87172 34.65354...	Pakistan	3564.9	2.251999e+08
26	BRA	POLYGON ((-53.37366 -33.76838, -53.65054 -33.2...	Brazil	4478.7	2.139934e+08
...	...	...	...	...	...
149	CZE	POLYGON ((15.01700 51.10667, 15.49097 50.78473...	Czechia	6113.0	1.070345e+07
120	GRC	MULTIPOLYGON (((26.29000 35.29999, 26.16500 35...	Greece	4272.7	1.066457e+07
107	SWE	POLYGON ((11.02737 58.85615, 11.46827 59.43239...	Sweden	5064.7	1.041581e+07
128	PRT	POLYGON ((-9.03482 41.88057, -8.67195 42.13469...	Portugal	5379.7	1.029942e+07
80	JOR	POLYGON ((35.54567 32.39399, 35.71992 32.70919...	Jordan	2290.1	1.026902e+07

87 rows × 5 columns

fig, ax = plt.subplots(figsize=(20,16))
earth.plot(ax=ax, color= "black")
pop_10269022.plot(ax=ax, column='2021_yield', legend=True, legend_kwds={"label": 'cereal yield for only the countries having a population between 10269022 and 1393409034', "orientation":"horizontal"}, cmap='Set1')
plt.show()

pop_10269022.explore(column='2021_yield',   # make choropleth based on "2021_yield" column
    tooltip=["Country Name_yield",'2021_yield','2021_pop'],  # show "Country Name_yield",'2021_yield','2021_pop' in tooltip (on hover)
    popup=True,  # show all values in popup (on click)
    tiles="openstreetmap",  # use "openstreetmap" tiles
    cmap="Set1",  # use "Set1" matplotlib colormap
    legend=True,                
    style_kwds=dict(color="black"), # use black outline
    zoom_control=True,
    #zoom_start = 1,
   #scrollWheelZoom=False
                    )

Make this Notebook Trusted to load map: File -> Trust Notebook

Brief Explanation:

From data frames and maps, I can observe that:

1. The United States of America (Gray) comes in first in this category with a cereal yield of 8268kg per hectare and a population of about 331 million.
2. The European countries of Belgium at 7906 cereal yield, the Netherlands at 7170kg per hectare, and France at 7170.9kg per hectare occupy the 2nd, 3rd, and 4th top positions, respectively.
3. Africa is represented in the 5th top position by Egypt, at 7132kg per hectare.
4. Africa is well represented at the bottom of the list of countries with the lowest cereal yield. The countries at the bottom are experiencing or have recently experienced civil wars and internal crises.
5. I noticed that India has the highest population in this category, but its crop yield is not close to the second country with the lowest population (Portugal). This is a policy area of concern for India to avoid the 'doom' of the Malthus theory of population growth.

Task 1.2.4

Plot (scatter or line plot) the percentage change in cereal yield from 2011 to 2021, for the country having the highest population in 2021. In this question, you must consider the cereal yield for each year between 2011 and 2021. Very briefly interpret the generated plot

earth_yield2.columns # inspect the parent dataframe for this question

Index(['Country Code', 'geometry', 'Country Name_yield', '2010_yield',
       '2011_yield', '2012_yield', '2013_yield', '2014_yield', '2015_yield',
       '2016_yield', '2017_yield', '2018_yield', '2019_yield', '2020_yield',
       '2021_yield', 'Country Name_pop', '2010_pop', '2011_pop', '2012_pop',
       '2013_pop', '2014_pop', '2015_pop', '2016_pop', '2017_pop', '2018_pop',
       '2019_pop', '2020_pop', '2021_pop'],
      dtype='object')

#filter for useful columns and cereal yeild from 2011 to 2021
earth_yield_2010_2021 = earth_yield2[['Country Code', 'geometry', 'Country Name_yield', '2021_pop', '2010_yield', '2011_yield',
       '2012_yield', '2013_yield', '2014_yield', '2015_yield', '2016_yield',
       '2017_yield', '2018_yield', '2019_yield', '2020_yield', '2021_yield']]

earth_yield_2010_2021

	Country Code	geometry	Country Name_yield	2021_pop	2010_yield	2011_yield	2012_yield	2013_yield	2014_yield	2015_yield	2016_yield	2017_yield	2018_yield	2019_yield	2020_yield	2021_yield
0	FJI	MULTIPOLYGON (((180.00000 -16.06713, 180.00000...	Fiji	902899.0	2871.1	2821.3	2423.4	2999.5	4475.4	3001.0	3000.9	3000.9	3002.2	3353.4	3665.7	3887.3
1	TZA	POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...	Tanzania	61498438.0	1647.9	1390.4	1314.8	1418.0	1524.8	1449.4	1570.4	1692.0	1938.5	1883.3	1698.4	1651.1
2	CAN	MULTIPOLYGON (((-122.84000 49.00000, -122.9742...	Canada	38246108.0	3501.1	3552.4	3456.3	4160.4	3647.0	3673.0	4239.1	4104.8	3914.7	4010.8	4095.4	3078.3
3	USA	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...	United States	331893745.0	6978.1	6803.5	5911.9	7300.9	7638.1	7430.1	8614.2	8281.3	8196.4	8006.1	8145.3	8268.0
4	KAZ	POLYGON ((87.35997 49.21498, 86.59878 48.54918...	Kazakhstan	19002586.0	804.1	1688.6	865.0	1164.9	1172.7	1278.1	1347.7	1355.0	1359.8	1154.5	1288.6	1049.0
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
163	BIH	POLYGON ((18.56000 42.65000, 17.67492 43.02856...	Bosnia and Herzegovina	3263459.0	3853.0	3724.9	3000.6	4026.9	3977.4	3812.1	5191.7	3732.1	5487.9	5354.7	6050.7	4422.1
164	MKD	POLYGON ((22.38053 42.32026, 22.88137 41.99930...	North Macedonia	2065092.0	3329.6	3502.1	2839.3	3381.3	3900.0	3051.1	3859.2	2807.5	3714.6	3554.0	3664.2	3537.8
165	SRB	POLYGON ((18.82982 45.90887, 18.82984 45.90888...	Serbia	6844078.0	4958.8	4751.4	3701.7	5157.8	5960.6	4787.3	6165.9	3967.9	6130.6	6126.9	6559.4	5768.6
166	MNE	POLYGON ((20.07070 42.58863, 19.80161 42.50009...	Montenegro	620173.0	3321.4	3305.7	2638.8	3770.5	3451.5	3146.5	3261.7	3288.0	3311.8	3166.8	3260.0	3330.7
167	TTO	POLYGON ((-61.68000 10.76000, -61.10500 10.890...	Trinidad and Tobago	1403374.0	1667.4	1639.5	1471.7	1610.5	1329.4	1110.4	1444.3	1627.5	1691.7	1529.7	1520.3	1588.1

164 rows × 16 columns

# I will use the .idxmax() method to get the highest population in 2021
high_pop_2021 = earth_yield_2010_2021.loc[earth_yield_2010_2021['2021_pop'].idxmax()]

high_pop_2021 # Highest population is China,  1412360000.0.

Country Code                                                        CHN
geometry              MULTIPOLYGON (((109.47520958866365 18.19770091...
Country Name_yield                                                China
2021_pop                                                   1412360000.0
2010_yield                                                       5526.1
2011_yield                                                       5709.4
2012_yield                                                       5827.1
2013_yield                                                       5894.1
2014_yield                                                       5893.2
2015_yield                                                       5985.7
2016_yield                                                       6017.6
2017_yield                                                       6111.3
2018_yield                                                       6125.4
2019_yield                                                       6265.9
2020_yield                                                       6314.2
2021_yield                                                       6320.8
Name: 136, dtype: object

high_pop_2021= high_pop_2021.iloc[4:] #I filter out the cereal yield from 2011 to 2021

high_pop_2021

2010_yield    5526.1
2011_yield    5709.4
2012_yield    5827.1
2013_yield    5894.1
2014_yield    5893.2
2015_yield    5985.7
2016_yield    6017.6
2017_yield    6111.3
2018_yield    6125.4
2019_yield    6265.9
2020_yield    6314.2
2021_yield    6320.8
Name: 136, dtype: object

high_pop_2021.pct_change()

2010_yield         NaN
2011_yield    0.033170
2012_yield    0.020615
2013_yield    0.011498
2014_yield   -0.000153
2015_yield    0.015696
2016_yield    0.005329
2017_yield    0.015571
2018_yield    0.002307
2019_yield    0.022937
2020_yield    0.007708
2021_yield    0.001045
Name: 136, dtype: float64

#Plotting percentage change in cereal yield from 2011 to 2021
high_pop_2021.pct_change().plot()

<Axes: >

Brief Explanation

1. The percentage change in cereal yield in China started at 0.033170% in 2011.
2. It declined steadily between 2011 and 2013, and in 2014 it went negative at -0.000153%.
3. There was quite a sharp rise in 2015 and a drop again in 2016.
4. It rose by almost the same proportion in 2017 as it declined in 2016. That is, the gradient of fall in 2016 is almost the same as the gradient of rise in 2017.
5. There is a sharp rise between 2018 and 2019.
6. From 2019 to 2021, the cereal yield went into free decline.
7. It eventually closed at 0.001045% after a series of falls and rises within the periods under consideration.

Task 1.2.5

Present a scatter plot between the mean population of each country and the mean cereal yield from the year 2011 until 2021. Very briefly interpret the generated plot, particularly looking for any correlation (if present) among the plotted variables. In this question, you must consider each year between 2011 and 2021 to find the mean population and mean cereal yield.

earth_yield2.columns

Index(['Country Code', 'geometry', 'Country Name_yield', '2010_yield',
       '2011_yield', '2012_yield', '2013_yield', '2014_yield', '2015_yield',
       '2016_yield', '2017_yield', '2018_yield', '2019_yield', '2020_yield',
       '2021_yield', 'Country Name_pop', '2010_pop', '2011_pop', '2012_pop',
       '2013_pop', '2014_pop', '2015_pop', '2016_pop', '2017_pop', '2018_pop',
       '2019_pop', '2020_pop', '2021_pop'],
      dtype='object')

I will drop the 2010 columns. They were used to explain percentage change in crop yield only.

earth_yield2.drop(['2010_yield', '2010_pop'], axis=1, inplace=True)

earth_yield2.columns

Index(['Country Code', 'geometry', 'Country Name_yield', '2011_yield',
       '2012_yield', '2013_yield', '2014_yield', '2015_yield', '2016_yield',
       '2017_yield', '2018_yield', '2019_yield', '2020_yield', '2021_yield',
       'Country Name_pop', '2011_pop', '2012_pop', '2013_pop', '2014_pop',
       '2015_pop', '2016_pop', '2017_pop', '2018_pop', '2019_pop', '2020_pop',
       '2021_pop'],
      dtype='object')

#Get the mean of the cereal yield and add mean yield column to the earth_yield data frame

earth_yield2['Mean_yield']=earth_yield2.iloc[:,3:13].mean(axis=1)

earth_yield2[:5]

	Country Code	geometry	Country Name_yield	2011_yield	2012_yield	2013_yield	2014_yield	2015_yield	2016_yield	2017_yield	...	2013_pop	2014_pop	2015_pop	2016_pop	2017_pop	2018_pop	2019_pop	2020_pop	2021_pop	Mean_yield
0	FJI	MULTIPOLYGON (((180.00000 -16.06713, 180.00000...	Fiji	2821.3	2423.4	2999.5	4475.4	3001.0	3000.9	3000.9	...	865602.0	866447.0	868632.0	872406.0	877460.0	883490.0	889955.0	896444.0	902899.0	3174.37
1	TZA	POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...	Tanzania	1390.4	1314.8	1418.0	1524.8	1449.4	1570.4	1692.0	...	48483132.0	49960563.0	51482638.0	53049231.0	54660345.0	56313444.0	58005461.0	59734213.0	61498438.0	1588.00
2	CAN	MULTIPOLYGON (((-122.84000 49.00000, -122.9742...	Canada	3552.4	3456.3	4160.4	3647.0	3673.0	4239.1	4104.8	...	35082954.0	35437435.0	35702908.0	36109487.0	36545236.0	37065084.0	37601230.0	38037204.0	38246108.0	3885.39
3	USA	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...	United States	6803.5	5911.9	7300.9	7638.1	7430.1	8614.2	8281.3	...	316059947.0	318386329.0	320738994.0	323071755.0	325122128.0	326838199.0	328329953.0	331501080.0	331893745.0	7632.78
4	KAZ	POLYGON ((87.35997 49.21498, 86.59878 48.54918...	Kazakhstan	1688.6	865.0	1164.9	1172.7	1278.1	1347.7	1355.0	...	17035551.0	17288285.0	17542806.0	17794055.0	18037776.0	18276452.0	18513673.0	18755666.0	19002586.0	1267.49

5 rows × 27 columns

earth_yield2.columns

Index(['Country Code', 'geometry', 'Country Name_yield', '2011_yield',
       '2012_yield', '2013_yield', '2014_yield', '2015_yield', '2016_yield',
       '2017_yield', '2018_yield', '2019_yield', '2020_yield', '2021_yield',
       'Country Name_pop', '2011_pop', '2012_pop', '2013_pop', '2014_pop',
       '2015_pop', '2016_pop', '2017_pop', '2018_pop', '2019_pop', '2020_pop',
       '2021_pop', 'Mean_yield'],
      dtype='object')

earth_yield2['Mean_pop']=earth_yield2.iloc[:,15:25].mean(axis=1)

earth_yield2[:2]

	Country Code	geometry	Country Name_yield	2011_yield	2012_yield	2013_yield	2014_yield	2015_yield	2016_yield	2017_yield	...	2014_pop	2015_pop	2016_pop	2017_pop	2018_pop	2019_pop	2020_pop	2021_pop	Mean_yield	Mean_pop
0	FJI	MULTIPOLYGON (((180.00000 -16.06713, 180.00000...	Fiji	2821.3	2423.4	2999.5	4475.4	3001.0	3000.9	3000.9	...	866447.0	868632.0	872406.0	877460.0	883490.0	889955.0	896444.0	902899.0	3174.37	874895.2
1	TZA	POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...	Tanzania	1390.4	1314.8	1418.0	1524.8	1449.4	1570.4	1692.0	...	49960563.0	51482638.0	53049231.0	54660345.0	56313444.0	58005461.0	59734213.0	61498438.0	1588.00	52441558.0

2 rows × 28 columns

select useful columns

earth_mean = earth_yield2[['geometry','Country Name_yield','Mean_yield', 'Mean_pop']]

earth_mean.sort_values(by='Mean_yield', ascending= False) #sort descending by mean yield

	geometry	Country Name_yield	Mean_yield	Mean_pop
81	POLYGON ((51.57952 24.24550, 51.75744 24.29407...	United Arab Emirates	25734.23	9390343.5
83	POLYGON ((47.97452 29.97582, 48.18319 29.53448...	Kuwait	12671.55	3819773.3
85	MULTIPOLYGON (((55.20834 22.70833, 55.23449 23...	Oman	12099.03	4286476.7
126	POLYGON ((6.15666 50.80372, 6.04307 50.12805, ...	Belgium	8760.31	11295471.1
127	POLYGON ((6.90514 53.48216, 7.09205 53.14404, ...	Netherlands	8385.30	17023700.7
...	...	...	...	...
86	MULTIPOLYGON (((167.21680 -15.89185, 167.84488...	Vanuatu	604.37	274734.8
13	POLYGON ((24.56737 8.22919, 23.80581 8.66632, ...	Sudan	600.36	39462148.6
46	POLYGON ((29.43219 -22.09131, 28.01724 -22.827...	Botswana	533.02	2160123.9
52	POLYGON ((14.85130 22.86295, 15.09689 21.30852...	Niger	498.37	20500747.4
47	POLYGON ((19.89577 -24.76779, 19.89473 -28.461...	Namibia	426.54	2341771.5

164 rows × 4 columns

earth_mean.sort_values(by='Mean_pop', ascending= False) #sort descending by mean population

	geometry	Country Name_yield	Mean_yield	Mean_pop
136	MULTIPOLYGON (((109.47521 18.19770, 108.65521 ...	China	6014.39	1.381980e+09
95	POLYGON ((97.32711 28.26158, 97.40256 27.88254...	India	3084.68	1.316492e+09
3	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...	United States	7632.78	3.215510e+08
7	MULTIPOLYGON (((141.00021 -2.60015, 141.01706 ...	Indonesia	5149.45	2.596911e+08
26	POLYGON ((-53.37366 -33.76838, -53.65054 -33.2...	Brazil	4786.45	2.052148e+08
...	...	...	...	...
145	POLYGON ((115.45071 5.44773, 115.40570 4.95523...	Brunei Darussalam	1657.45	4.165801e+05
18	MULTIPOLYGON (((-78.98000 26.79000, -78.51000 ...	Bahamas, The	7870.76	3.763192e+05
36	POLYGON ((-89.14308 17.80832, -89.15091 17.955...	Belize	3462.48	3.643453e+05
86	MULTIPOLYGON (((167.21680 -15.89185, 167.84488...	Vanuatu	604.37	2.747348e+05
131	POLYGON ((165.77999 -21.08000, 166.59999 -21.7...	New Caledonia	5758.90	2.667010e+05

164 rows × 4 columns

INVESTIGATING RELATIONSHIP

I will have a working hypothesis:

Ho: There is no significant relationship between the mean population and the mean yield.

Please note that I have taken mean yield as my dependent variable and placed it on the Y-axis and the independent variable, population, on the X-axis.

earth_mean.plot(kind='scatter', x='Mean_pop', y='Mean_yield')

<Axes: xlabel='Mean_pop', ylabel='Mean_yield'>

I will plot this with plotly.express for better view and easier explanation

import plotly.express as px

fig = px.scatter(earth_mean,x='Mean_pop', y='Mean_yield', color= 'Country Name_yield')
fig.show()

Pionts are too clustered at the bottom-right, I will cut off the outliers of population (at 3 million) and mean yield (at 10,000)

cut_outliers= earth_mean.loc[(earth_mean.Mean_pop<300000000) & (earth_mean.Mean_yield<10000)]

fig = px.scatter(cut_outliers,x='Mean_pop', y='Mean_yield', color= 'Country Name_yield')
fig.show()

Points are easier to view now, I can accept my null hypothesis at this point. However, I shall attempt a spatial view of possible relationships

fig, ax = plt.subplots(figsize=(20,16))
earth.plot(ax=ax, color= "black")
earth_mean.plot(ax=ax, column='Mean_yield', legend=True, legend_kwds={"label":'Mean cereal yield of countries between  2011 and 2021', "orientation":"horizontal"}, cmap='Set1')
plt.show()

earth_mean.explore(column='Mean_yield',   # make choropleth based on "BoroName" column
    tooltip=['Country Name_yield','Mean_yield','Mean_pop',],  # show "BoroName" value in tooltip (on hover)
    popup=True,  # show all values in popup (on click)
    tiles="openstreetmap",  # use "CartoDB positron" tiles
    cmap="Set1",  # use "Set1" matplotlib colormap
    legend=True,                
    style_kwds=dict(color="black"), # use black outline
    zoom_control=True,
    zoom_start = 1,
                  )

Make this Notebook Trusted to load map: File -> Trust Notebook

Brief Explanation

1. I did not find a significant relationship in values between the mean population and the mean cereal yield.
2. There is, however, a spatial relationship. It is regionalized.
3. Africa is the worst-performing region in cereal yield.
4. No northern American country is at the bottom in cereal yield.
5. Bolivia is the only South American country at the bottom in cereal production within the period under consideration.
6. The United Arab Emirates (popularly called Dubai) has maintained a consistent lead in cereal yield over the years.
7. Kuwait and Oman are the 2nd and 3rd best-performing countries in cereal yield, respectively.
8. Belgium and the Netherlands are the 4th and 5th best-performing countries in the world, and they are the 1st and 2nd in Europe, respectively.
9. New Caledonia, despite its low population, is a top-performing country in cereal yield per hectare.
10. African countries need to increase their cereal yield production to prevent extreme hunger and increase their balance of trade..

TASK 2: Gospatial Sentiment Analysis Using Social Media Data

INTRODUCTION

In this part, I will apply geospatial sentiment analysis to Twitter data using the Python library, TextBlob. I am provided with a dataset consisting of tweets relevant to cryptocurrency

Task 2.1: Data Pre-processing

Instruction

Using a set of suitable Python libraries, randomly retrieve 500 tweets where user locations are available. You should also filter out the irrelevant characters, symbols, hashtags, URLs etc. from the tweets to avoid any possible masking of the actual sentiment associated with the tweets. From this point onward you should use the processed tweet data for all the subsequent analyses.

I will start by installing textblob

TextBlob is a Python library for processing textual data. It provides a simple API for diving into common natural language processing such as sentiment analysis. It is built on top of NLTK and Pattern (TextBlob: Simplified Text Processing — TextBlob 0.16.0 Documentation, 2023; Muhammad, 2023)

!pip install textblob

Requirement already satisfied: textblob in c:\users\user\anaconda3\lib\site-packages (0.17.1)
Requirement already satisfied: nltk>=3.1 in c:\users\user\anaconda3\lib\site-packages (from textblob) (3.8.1)
Requirement already satisfied: click in c:\users\user\anaconda3\lib\site-packages (from nltk>=3.1->textblob) (8.0.4)
Requirement already satisfied: joblib in c:\users\user\anaconda3\lib\site-packages (from nltk>=3.1->textblob) (1.2.0)
Requirement already satisfied: regex>=2021.8.3 in c:\users\user\anaconda3\lib\site-packages (from nltk>=3.1->textblob) (2022.7.9)
Requirement already satisfied: tqdm in c:\users\user\anaconda3\lib\site-packages (from nltk>=3.1->textblob) (4.65.0)
Requirement already satisfied: colorama in c:\users\user\anaconda3\lib\site-packages (from click->nltk>=3.1->textblob) (0.4.6)

Next will be to import my python libraries. Matplotlib is a python visualition tool. Pandas will help me create a dataframe and manage the dataframe

from textblob import TextBlob
import matplotlib.pyplot as plt
import pandas as pd

The bitcoin tweet file is huge with more than 4million rows. I beg your indulgence take a sample of 1 million rows. One million is a fair representation of the total sample

bitcoin =pd.read_csv(r"Bitcoin_tweets.csv", nrows=1000000) 

C:\Users\User\AppData\Local\Temp\ipykernel_8832\894283730.py:1: DtypeWarning:

Columns (5,7,8,12) have mixed types. Specify dtype option on import or set low_memory=False.

bitcoin

	user_name	user_location	user_description	user_created	user_followers	user_friends	user_favourites	user_verified	date	text	hashtags	source	is_retweet
0	DeSota Wilson	Atlanta, GA	Biz Consultant, real estate, fintech, startups...	39929.836910	8534	7605	4838.0	FALSE	44237.9993518519	Blue Ridge Bank shares halted by NYSE after #b...	['bitcoin']	Twitter Web App	False
1	CryptoND	NaN	😎 BITCOINLIVE is a Dutch platform aimed at inf...	43755.841782	6769	1532	25483.0	FALSE	44237.9991666667	😎 Today, that's this #Thursday, we will do a "...	['Thursday', 'Btc', 'wallet', 'security']	Twitter for Android	False
2	Tdlmatias	London, England	IM Academy : The best #forex, #SelfEducation, ...	41953.451817	128	332	924.0	FALSE	44237.9963888889	Guys evening, I have read this article about B...	NaN	Twitter Web App	False
3	Crypto is the future	NaN	I will post a lot of buying signals for BTC tr...	43736.700139	625	129	14.0	FALSE	44237.9962152778	$BTC A big chance in a billion! Price: \487264...	['Bitcoin', 'FX', 'BTC', 'crypto']	dlvr.it	False
4	Alex Kirchmaier 🇦🇹🇸🇪 #FactsSuperspreader	Europa	Co-founder @RENJERJerky | Forbes 30Under30 | I...	42403.552720	1249	1472	10482.0	FALSE	44237.9959027778	This network is secured by 9 508 nodes as of t...	['BTC']	Twitter Web App	False
...	...	...	...	...	...	...	...	...	...	...	...	...	...
999995	Bob Lji ∞/21 🟠	Interlaken, Schweiz	DYOR + DCA + HODL #BITCOIN NoAlts+NoCryptos+No...	44355.652176	391	1558	1243.0	False	44423.537072	#wtfhappenedin1971? \n#Bitcoin will fix this!	['wtfhappenedin1971', 'Bitcoin']	Twitter for iPhone	False
999996	MoonTrip	Blockchain	Crypto Enthusiast	44289.838113	384	491	4771.0	False	44423.537049	@CryptoNadine @BabyArabiaBSC #babyarabia massi...	['babyarabia', 'bitcoin', 'dev', 'sundayvibes'...	Twitter for Android	False
999997	I like Tea and #BSV	NaN	I like Tea and Buying #BSV	44315.271470	0	0	2.0	False	44423.536944	@mcuban Are you serious or part of the scam??\...	['BSV', 'BITCOIN', 'BSV', 'greentechnology']	Twitter for Android	False
999998	The Blockchain Advisor™	Illinois, USA	Bridging the Gap Between Traditional Investing...	40259.683646	970	854	9134.0	False	44423.536725	Why is #Bitcoin going to ascend to a global re...	['Bitcoin']	Twitter Web App	False
999999	redforthebest	NaN	Creator of #CosmoHeads & #MintingClassics & #...	44313.291424	2357	4571	1448.0	False	44423.536528	#Bitcoin in #QRthings Collection\n(Only 0.03 e...	['Bitcoin', 'QRthings', 'Crypto', 'Ethereum', ...	Twitter for Android	False

1000000 rows × 13 columns

TextBlob is a Python library for processing textual data. It provides a simple API for diving into common natural language processing such as sentiment analysis. It is built on top of NLTK and Pattern (TextBlob: Simplified Text Processing — TextBlob 0.16.0 Documentation, 2023; Muhammad, 2023)

bitcoin= bitcoin.sample(n = 1500) # My first selection will be 1500 this is to compensate for missing values as null vlues will be drop

bitcoin

	user_name	user_location	user_description	user_created	user_followers	user_friends	user_favourites	user_verified	date	text	hashtags	source	is_retweet
904542	probtclife ⚡	NaN	#Bitcoin	44203.483310	27	142	4992.0	False	44427.516875	@GaryGensler @FoxBusiness #Bitcoin fixes this	['Bitcoin']	Twitter Web App	False
222839	J***** T******	Mainz	Get free $DFI worth 30$ @cakedefi, use the fol...	44107.647488	44	194	3930.0	False	44368.595752	@CaptainCryptoHD @Cryptanzee Thanks for the qu...	['Bitcoin']	Twitter for Android	False
75023	Maximus Monius	NaN	Crypto News, Technical Analysis, Tips & Tricks...	44246.658889	72	130	102.0	False	44310.680347	#BITCOIN UPDATE - April 24th\n\nWe seem to be ...	['BITCOIN', 'BTC', 'ALTSEASON']	Twitter Web App	False
951672	Rahul Singh Sirohi	Ghaziabad, India	Advisor & Developer in #Crypto Industry Since ...	42906.352164	1406	1980	5134.0	False	44425.657338	When #Bitcoin Crashed & goes to #Bearish M...	['Bitcoin', 'Bearish', 'BTC', 'Crypto', 'AskTo...	Twitter for Android	False
339824	JcL.Rheendy	NaN	NaN	44039.657245	40	1357	594.0	False	44379.606065	@WSB_WallStreet one of the very big giveaways ...	['giveaway', 'bitcoin', 'USDC', 'ETHEREUM', 'T...	Twitter for Android	False
...	...	...	...	...	...	...	...	...	...	...	...	...	...
912730	BitcoinAgile	Matter Doesn't Matter	Breaking News. Bitcoin, Blockchain & Beyond. #...	41646.994977	59940	13883	8737.0	False	44427.132847	TA: #bitcoin Turns Red, What Could Trigger Mor...	['bitcoin', 'btcusd', 'btcusdt', 'xbtusd']	bitcoinagile	False
347606	VaxBLR	Bengaluru, India	Hourly updates on FREE and PAID 18+ and 45+ va...	44368.364282	12	0	0.0	False	44400.375208	45+ #URBAN #Bengaluru #CovidVaccine Availabili...	['URBAN', 'Bengaluru', 'CovidVaccine', 'COVISH...	VaxBlr	False
426243	𝕍ℝaj 🇮🇳	Mumbai	Beating Inflation by Equities. Penchant for Pa...	40552.570046	117	917	4369.0	False	44399.662593	@ramrastogi @LinkedIn Sir I think inexorable c...	['Bitcoin']	Twitter for Android	False
366435	thank u, next	Greece	NaN	40998.724236	699	1954	96384.0	False	44400.478669	The opening ceremony is so boring like people ...	['OlympicGames']	Twitter for iPhone	False
386290	Mantas	NaN	Crypto enthusiast 🪙🚀🌕🐂	43099.398368	16	175	999.0	False	44401.31206	Bullish. #BTC #Bitcoin	['BTC', 'Bitcoin']	Twitter for iPhone	False

1500 rows × 13 columns

bitcoin.isna().sum()

user_name             0
user_location       710
user_description    163
user_created          0
user_followers        0
user_friends          0
user_favourites       0
user_verified         0
date                  0
text                  0
hashtags             31
source                3
is_retweet            0
dtype: int64

bitcoin.dropna(inplace= True)

bitcoin.isna().sum()

user_name           0
user_location       0
user_description    0
user_created        0
user_followers      0
user_friends        0
user_favourites     0
user_verified       0
date                0
text                0
hashtags            0
source              0
is_retweet          0
dtype: int64

bitcoin.shape

(737, 13)

bitcoin #inspect DF

	user_name	user_location	user_description	user_created	user_followers	user_friends	user_favourites	user_verified	date	text	hashtags	source	is_retweet
222839	J***** T******	Mainz	Get free $DFI worth 30$ @cakedefi, use the fol...	44107.647488	44	194	3930.0	False	44368.595752	@CaptainCryptoHD @Cryptanzee Thanks for the qu...	['Bitcoin']	Twitter for Android	False
951672	Rahul Singh Sirohi	Ghaziabad, India	Advisor & Developer in #Crypto Industry Since ...	42906.352164	1406	1980	5134.0	False	44425.657338	When #Bitcoin Crashed & goes to #Bearish M...	['Bitcoin', 'Bearish', 'BTC', 'Crypto', 'AskTo...	Twitter for Android	False
788279	🔔system'cRe5520'	AWS eu-west-1a Ireland Region	The channel breakout trading strategy bot for ...	41062.324120	158	16	63.0	False	44417.625035	strategy: 5010HL1h atr20d: 2196.69\n\n09 Aug 2...	['BTC', 'BitMEX']	system'cRe5520'	False
456359	Cryptorphic	Moon	#Bitcoin Certified Technical Analyst I Margin ...	43283.686458	4176	33	249.0	False	44398.774907	IMHO Do not long or Short anything now with le...	['Crypto', 'Btc', 'Bitcoin']	Twitter for iPhone	False
741240	DoopieCash®	The Hague, The Netherlands	▫ Professional Crypto & FX Trader ▫ Technical ...	43203.412604	21365	209	4714.0	False	44417.600694	Time to melt 🔥🔥🔥\n\n#Bitcoin on fire\n\n#BTC ...	['Bitcoin', 'BTC', 'crypto']	Twitter Web App	False
...	...	...	...	...	...	...	...	...	...	...	...	...	...
356131	JJ	Raleigh, NC	nobody really cares what I put here.	40875.925961	306	462	8626.0	False	44393.935093	So many dumb replies. The #CovidVaccine does n...	['CovidVaccine', 'VaccineEducation']	Twitter Web App	False
665361	Andre phill	Ontario, Canada	Masters at business Administration-MBA at Ohio...	44403.469086	8	49	12.0	FALSE	44405.1377083333	@APompliano Trust has an unlimited value, The ...	['BitcoinCash', 'money', 'dollar', 'btc']	Twitter for iPhone	False
912730	BitcoinAgile	Matter Doesn't Matter	Breaking News. Bitcoin, Blockchain & Beyond. #...	41646.994977	59940	13883	8737.0	False	44427.132847	TA: #bitcoin Turns Red, What Could Trigger Mor...	['bitcoin', 'btcusd', 'btcusdt', 'xbtusd']	bitcoinagile	False
347606	VaxBLR	Bengaluru, India	Hourly updates on FREE and PAID 18+ and 45+ va...	44368.364282	12	0	0.0	False	44400.375208	45+ #URBAN #Bengaluru #CovidVaccine Availabili...	['URBAN', 'Bengaluru', 'CovidVaccine', 'COVISH...	VaxBlr	False
426243	𝕍ℝaj 🇮🇳	Mumbai	Beating Inflation by Equities. Penchant for Pa...	40552.570046	117	917	4369.0	False	44399.662593	@ramrastogi @LinkedIn Sir I think inexorable c...	['Bitcoin']	Twitter for Android	False

737 rows × 13 columns

bitcoin['user_location'] #inspect location

222839                            Mainz
951672                 Ghaziabad, India
788279    AWS eu-west-1a Ireland Region
456359                             Moon
741240       The Hague, The Netherlands
                      ...              
356131                      Raleigh, NC
665361                  Ontario, Canada
912730            Matter Doesn't Matter
347606                 Bengaluru, India
426243                          Mumbai 
Name: user_location, Length: 737, dtype: object

I will filter out columns that are useful to my study

bitcoin.columns

Index(['user_name', 'user_location', 'user_description', 'user_created',
       'user_followers', 'user_friends', 'user_favourites', 'user_verified',
       'date', 'text', 'hashtags', 'source', 'is_retweet'],
      dtype='object')

bitcoin = bitcoin[['user_location', 'text', 'user_followers']]

bitcoin

	user_location	text	user_followers
222839	Mainz	@CaptainCryptoHD @Cryptanzee Thanks for the qu...	44
951672	Ghaziabad, India	When #Bitcoin Crashed & goes to #Bearish M...	1406
788279	AWS eu-west-1a Ireland Region	strategy: 5010HL1h atr20d: 2196.69\n\n09 Aug 2...	158
456359	Moon	IMHO Do not long or Short anything now with le...	4176
741240	The Hague, The Netherlands	Time to melt 🔥🔥🔥\n\n#Bitcoin on fire\n\n#BTC ...	21365
...	...	...	...
356131	Raleigh, NC	So many dumb replies. The #CovidVaccine does n...	306
665361	Ontario, Canada	@APompliano Trust has an unlimited value, The ...	8
912730	Matter Doesn't Matter	TA: #bitcoin Turns Red, What Could Trigger Mor...	59940
347606	Bengaluru, India	45+ #URBAN #Bengaluru #CovidVaccine Availabili...	12
426243	Mumbai	@ramrastogi @LinkedIn Sir I think inexorable c...	117

737 rows × 3 columns

Task 2.1: Data Pre-processing

Using a set of suitable Python libraries, randomly retrieve 500 tweets where user locations are available. You should also filter out the irrelevant characters, symbols, hashtags, URLs etc. from the tweets to avoid any possible masking of the actual sentiment associated with the tweets. From this point onward you should use the processed tweet data for all the subsequent analyses.

import re

I am going to define a function that identifies and removes irrelevant characters from the tweets. This is the essence of the 'import re' statement above. A regular expression (or RE) specifies a set of strings that match it; the functions in this module let you check if a particular string matches a given regular expression (Re: Regular Expression Operations, 2023).

def remove_rt(x): return re.sub('RT @\w+: ', " ", x)
 
def rt(x): return re.sub(
    "(@[A-Za-z0-9]+)|([^0-9A-Za-z \t])|(\w+:\/\/\S+)", " ", x)
 
bitcoin["text"] = bitcoin.text.map(remove_rt).map(rt)
bitcoin["text"] = bitcoin.text.str.lower()

C:\Users\User\AppData\Local\Temp\ipykernel_8832\2926551215.py:6: SettingWithCopyWarning:


A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy

C:\Users\User\AppData\Local\Temp\ipykernel_8832\2926551215.py:7: SettingWithCopyWarning:


A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy

bitcoin #inspect dataset

	user_location	text	user_followers
222839	Mainz	thanks for the question my favourite one...	44
951672	Ghaziabad, India	when bitcoin crashed amp goes to bearish m...	1406
788279	AWS eu-west-1a Ireland Region	strategy 5010hl1h atr20d 2196 69 09 aug 202...	158
456359	Moon	imho do not long or short anything now with le...	4176
741240	The Hague, The Netherlands	time to melt bitcoin on fire btc cry...	21365
...	...	...	...
356131	Raleigh, NC	so many dumb replies the covidvaccine does n...	306
665361	Ontario, Canada	trust has an unlimited value the father of ...	8
912730	Matter Doesn't Matter	ta bitcoin turns red what could trigger mor...	59940
347606	Bengaluru, India	45 urban bengaluru covidvaccine availabili...	12
426243	Mumbai	sir i think inexorable certainty that cent...	117

737 rows × 3 columns

print(bitcoin['user_location'].value_counts()) #inspect before geocoding

user_location
Bay Area, CA                 25
United States                13
Global                       12
London, England              12
Australia                    11
                             ..
Croatia                       1
Bangladesh...... Khulna       1
Lewes, DE                     1
florida                       1
Mumbai                        1
Name: count, Length: 469, dtype: int64

Task 2.2: Geocoding

Geocode on all the 500 tweets retrieved and filtered in the previous step. To perform geocoding, you must be using a Python-based tool. Once the geocoding is performed then augment the tweet data set with two extra columns. One column should contain latitude and the other one should contain longitude information corresponding to a tweet

geopy is a Python client for several popular geocoding web services (Geopy, 2023),

geopy. (2023, November 23). PyPI. https://pypi.org/project/geopy/

pip install geopy 

Requirement already satisfied: geopy in c:\users\user\anaconda3\lib\site-packages (2.4.1)Note: you may need to restart the kernel to use updated packages.

Requirement already satisfied: geographiclib<3,>=1.52 in c:\users\user\anaconda3\lib\site-packages (from geopy) (2.0)

# Import Nominatim
from geopy.geocoders import Nominatim

geolocator = Nominatim(user_agent="Assessment")

from geopy.extra.rate_limiter import RateLimiter

I am going to duplicate the user_location column but call it locations. The essence is to have a column that keeps the original state of the user location ater applying geocoding to the fataframe

bitcoin['locations'] = bitcoin['user_location'] 

C:\Users\User\AppData\Local\Temp\ipykernel_8832\3072469440.py:1: SettingWithCopyWarning:


A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy

bitcoin

	user_location	text	user_followers	locations
222839	Mainz	thanks for the question my favourite one...	44	Mainz
951672	Ghaziabad, India	when bitcoin crashed amp goes to bearish m...	1406	Ghaziabad, India
788279	AWS eu-west-1a Ireland Region	strategy 5010hl1h atr20d 2196 69 09 aug 202...	158	AWS eu-west-1a Ireland Region
456359	Moon	imho do not long or short anything now with le...	4176	Moon
741240	The Hague, The Netherlands	time to melt bitcoin on fire btc cry...	21365	The Hague, The Netherlands
...	...	...	...	...
356131	Raleigh, NC	so many dumb replies the covidvaccine does n...	306	Raleigh, NC
665361	Ontario, Canada	trust has an unlimited value the father of ...	8	Ontario, Canada
912730	Matter Doesn't Matter	ta bitcoin turns red what could trigger mor...	59940	Matter Doesn't Matter
347606	Bengaluru, India	45 urban bengaluru covidvaccine availabili...	12	Bengaluru, India
426243	Mumbai	sir i think inexorable certainty that cent...	117	Mumbai

737 rows × 4 columns

geocode = RateLimiter(geolocator.geocode, min_delay_seconds=1)

bitcoin['user_location'] = bitcoin['locations'].apply(geocode)
bitcoin['latitude'] = bitcoin['user_location'].apply(lambda x: x.latitude if x else None)
bitcoin['longitude'] = bitcoin['user_location'].apply(lambda x: x.longitude if x else None)

bitcoin

print(bitcoin['locations'].value_counts())

bitcoin.dropna(inplace = True)

bitcoin.isna().sum()

bitcoin

Next step will be to take out a sample 500 at this point as all tweets in my bitcoin2 dataframe have location

bitcoin2=bitcoin.sample(n=500) 

bitcoin2

bitcoin2.to_csv('bitcoin_final.csv') #save dataframe as a csv file for future reference

bitcoin2 =gpd.read_file('bitcoin_final.csv')

bitcoin2

	field_1	user_location	text	user_followers	locations	latitude	longitude	geometry
0	714437	The Moon, Sylacauga, Talladega County, Alabama...	talked to about stacks stx and hopes the...	39	The Moon	33.1584497	-86.2385846	None
1	280261	Lisboa, Portugal	i m so f tired of ignorant american phd econom...	1614	Lisbon, Portugal	38.7077507	-9.1365919	None
2	336653	Muhu, Saare maakond, Eesti	long bitcoin	43	Moon	58.5959044	23.21964608602439	None
3	872349	Global, 81, Barber Greene Road, Don Mills, Don...	bitcoin bull cathie wood attracts big short...	77240	Global	43.7283874	-79.34914879325001	None
4	109405	Canada	bitcoin price in us dollar btc usd btcusd ...	250	Canada	61.0666922	-107.991707	None
...	...	...	...	...	...	...	...	...
495	898076	Hyderabad, Bahadurpura mandal, Hyderabad Distr...	freedom 35 official bsc bitcoin via	112	Hyderabad, India	17.360589	78.4740613	None
496	951645	b'424c4f434b434841494e2c20d09dd0b0d0b1d0b5d180...	bitcoin btc i don t like these lower highs...	21254	#blockchain	44.6465984	34.4007341	None
497	7974	Sydney, Council of the City of Sydney, New Sou...	what goes up must comes down hard question ...	26	Sydney, New South Wales	-33.8698439	151.2082848	None
498	788995	Montréal, Agglomération de Montréal, MontrÃ...	am i the only one that likes watching the sats...	135	Montréal, Québec	45.5031824	-73.5698065	None
499	62652	Laguna Beach, Orange County, California, Unite...	lamb is a fast safe and scalable blockchai...	3784	Laguna Beach, CA	33.5426975	-117.785366	None

500 rows × 8 columns

bitcoin2.reset_index(drop=True, inplace=True) #reset index for a nice looking DF

bitcoin2 #inspect

	field_1	user_location	text	user_followers	locations	latitude	longitude	geometry
0	714437	The Moon, Sylacauga, Talladega County, Alabama...	talked to about stacks stx and hopes the...	39	The Moon	33.1584497	-86.2385846	None
1	280261	Lisboa, Portugal	i m so f tired of ignorant american phd econom...	1614	Lisbon, Portugal	38.7077507	-9.1365919	None
2	336653	Muhu, Saare maakond, Eesti	long bitcoin	43	Moon	58.5959044	23.21964608602439	None
3	872349	Global, 81, Barber Greene Road, Don Mills, Don...	bitcoin bull cathie wood attracts big short...	77240	Global	43.7283874	-79.34914879325001	None
4	109405	Canada	bitcoin price in us dollar btc usd btcusd ...	250	Canada	61.0666922	-107.991707	None
...	...	...	...	...	...	...	...	...
495	898076	Hyderabad, Bahadurpura mandal, Hyderabad Distr...	freedom 35 official bsc bitcoin via	112	Hyderabad, India	17.360589	78.4740613	None
496	951645	b'424c4f434b434841494e2c20d09dd0b0d0b1d0b5d180...	bitcoin btc i don t like these lower highs...	21254	#blockchain	44.6465984	34.4007341	None
497	7974	Sydney, Council of the City of Sydney, New Sou...	what goes up must comes down hard question ...	26	Sydney, New South Wales	-33.8698439	151.2082848	None
498	788995	Montréal, Agglomération de Montréal, MontrÃ...	am i the only one that likes watching the sats...	135	Montréal, Québec	45.5031824	-73.5698065	None
499	62652	Laguna Beach, Orange County, California, Unite...	lamb is a fast safe and scalable blockchai...	3784	Laguna Beach, CA	33.5426975	-117.785366	None

500 rows × 8 columns

Next step will be to convert longitude and latitude to geometry and assign it to "EPSG:4326" coordinate reference system Code Source: GeoPandas Documentation (2023)

bitcoin_tweet = gpd.GeoDataFrame(bitcoin2, geometry=gpd.points_from_xy(bitcoin2.longitude, bitcoin2.latitude), crs="EPSG:4326"
)

bitcoin_tweet

	field_1	user_location	text	user_followers	locations	latitude	longitude	geometry
0	714437	The Moon, Sylacauga, Talladega County, Alabama...	talked to about stacks stx and hopes the...	39	The Moon	33.1584497	-86.2385846	POINT (-86.23858 33.15845)
1	280261	Lisboa, Portugal	i m so f tired of ignorant american phd econom...	1614	Lisbon, Portugal	38.7077507	-9.1365919	POINT (-9.13659 38.70775)
2	336653	Muhu, Saare maakond, Eesti	long bitcoin	43	Moon	58.5959044	23.21964608602439	POINT (23.21965 58.59590)
3	872349	Global, 81, Barber Greene Road, Don Mills, Don...	bitcoin bull cathie wood attracts big short...	77240	Global	43.7283874	-79.34914879325001	POINT (-79.34915 43.72839)
4	109405	Canada	bitcoin price in us dollar btc usd btcusd ...	250	Canada	61.0666922	-107.991707	POINT (-107.99171 61.06669)
...	...	...	...	...	...	...	...	...
495	898076	Hyderabad, Bahadurpura mandal, Hyderabad Distr...	freedom 35 official bsc bitcoin via	112	Hyderabad, India	17.360589	78.4740613	POINT (78.47406 17.36059)
496	951645	b'424c4f434b434841494e2c20d09dd0b0d0b1d0b5d180...	bitcoin btc i don t like these lower highs...	21254	#blockchain	44.6465984	34.4007341	POINT (34.40073 44.64660)
497	7974	Sydney, Council of the City of Sydney, New Sou...	what goes up must comes down hard question ...	26	Sydney, New South Wales	-33.8698439	151.2082848	POINT (151.20828 -33.86984)
498	788995	Montréal, Agglomération de Montréal, MontrÃ...	am i the only one that likes watching the sats...	135	Montréal, Québec	45.5031824	-73.5698065	POINT (-73.56981 45.50318)
499	62652	Laguna Beach, Orange County, California, Unite...	lamb is a fast safe and scalable blockchai...	3784	Laguna Beach, CA	33.5426975	-117.785366	POINT (-117.78537 33.54270)

500 rows × 8 columns

Task 2.3 Polarity analysis

Calculate the polarity values of all the tweets. For a given geographical location, if you have more than one tweet then find the average polarity value taking into consideration all the tweets generated from the same location. Using a suitable plot type (such as a geographical map), perform a geospatial visualisation of the polarities corresponding to all the tweets. Whilst you are free to choose a plot type, the visualisation must be clear and easy to understand/interpret.

#define a funtion that gets text polarity
def getTextPolarity(txt):
    return TextBlob(txt).sentiment.polarity

bitcoin_tweet['polarity'] = bitcoin_tweet['text'].apply(getTextPolarity)
bitcoin_tweet.head()

	field_1	user_location	text	user_followers	locations	latitude	longitude	geometry	polarity
0	714437	The Moon, Sylacauga, Talladega County, Alabama...	talked to about stacks stx and hopes the...	39	The Moon	33.1584497	-86.2385846	POINT (-86.23858 33.15845)	-0.125000
1	280261	Lisboa, Portugal	i m so f tired of ignorant american phd econom...	1614	Lisbon, Portugal	38.7077507	-9.1365919	POINT (-9.13659 38.70775)	-0.133333
2	336653	Muhu, Saare maakond, Eesti	long bitcoin	43	Moon	58.5959044	23.21964608602439	POINT (23.21965 58.59590)	-0.050000
3	872349	Global, 81, Barber Greene Road, Don Mills, Don...	bitcoin bull cathie wood attracts big short...	77240	Global	43.7283874	-79.34914879325001	POINT (-79.34915 43.72839)	0.000000
4	109405	Canada	bitcoin price in us dollar btc usd btcusd ...	250	Canada	61.0666922	-107.991707	POINT (-107.99171 61.06669)	0.165000

import seaborn as sns #This is library built ontop matplotlib, it aids visualisation

sns.histplot(x='polarity', data= bitcoin_tweet, color='green') #Plot polarity values count with sns

<Axes: xlabel='polarity', ylabel='Count'>

Define a funtion that labels the polarity values, (Oluyale, 2023).

def  definepolarity(x):
  if x > 0.00:
    return "Positive" 
  elif x < 0.00: 
    return "Negative" 
  elif x == 0: 
    return "Neutral" 

Apply the defined function and create another column for polarity label

bitcoin_tweet["polarity_label"] = bitcoin_tweet["polarity"].apply(definepolarity) 

bitcoin_tweet #inspect

	field_1	user_location	text	user_followers	locations	latitude	longitude	geometry	polarity	polarity_label
0	714437	The Moon, Sylacauga, Talladega County, Alabama...	talked to about stacks stx and hopes the...	39	The Moon	33.1584497	-86.2385846	POINT (-86.23858 33.15845)	-0.125000	Negative
1	280261	Lisboa, Portugal	i m so f tired of ignorant american phd econom...	1614	Lisbon, Portugal	38.7077507	-9.1365919	POINT (-9.13659 38.70775)	-0.133333	Negative
2	336653	Muhu, Saare maakond, Eesti	long bitcoin	43	Moon	58.5959044	23.21964608602439	POINT (23.21965 58.59590)	-0.050000	Negative
3	872349	Global, 81, Barber Greene Road, Don Mills, Don...	bitcoin bull cathie wood attracts big short...	77240	Global	43.7283874	-79.34914879325001	POINT (-79.34915 43.72839)	0.000000	Neutral
4	109405	Canada	bitcoin price in us dollar btc usd btcusd ...	250	Canada	61.0666922	-107.991707	POINT (-107.99171 61.06669)	0.165000	Positive
...	...	...	...	...	...	...	...	...	...	...
495	898076	Hyderabad, Bahadurpura mandal, Hyderabad Distr...	freedom 35 official bsc bitcoin via	112	Hyderabad, India	17.360589	78.4740613	POINT (78.47406 17.36059)	0.000000	Neutral
496	951645	b'424c4f434b434841494e2c20d09dd0b0d0b1d0b5d180...	bitcoin btc i don t like these lower highs...	21254	#blockchain	44.6465984	34.4007341	POINT (34.40073 44.64660)	0.000000	Neutral
497	7974	Sydney, Council of the City of Sydney, New Sou...	what goes up must comes down hard question ...	26	Sydney, New South Wales	-33.8698439	151.2082848	POINT (151.20828 -33.86984)	-0.223611	Negative
498	788995	Montréal, Agglomération de Montréal, MontrÃ...	am i the only one that likes watching the sats...	135	Montréal, Québec	45.5031824	-73.5698065	POINT (-73.56981 45.50318)	0.250000	Positive
499	62652	Laguna Beach, Orange County, California, Unite...	lamb is a fast safe and scalable blockchai...	3784	Laguna Beach, CA	33.5426975	-117.785366	POINT (-117.78537 33.54270)	0.350000	Positive

500 rows × 10 columns

#Plot polarity label count with sns

plt.figure(figsize=(12,4))
sns.countplot(x='polarity_label', data=bitcoin_tweet)
plt.title('Polarity Label Count')
plt.show()

bitcoin_tweet.columns

Index(['field_1', 'user_location', 'text', 'user_followers', 'locations',
       'latitude', 'longitude', 'geometry', 'polarity', 'polarity_label'],
      dtype='object')

Filter the useful columns for map visualization

bitcoin_tweet2= bitcoin_tweet[['geometry', 'polarity', 'polarity_label']]

Plotting Polarity

Polarity Values

fig, ax = plt.subplots(figsize=(20,16))
earth.plot(ax=ax, color= "gray")
bitcoin_tweet2.plot(ax=ax, column='polarity', legend=True, legend_kwds={"label": 'Polarity Values Spread of Bitcoin Tweets', "orientation":"horizontal"}, cmap='Set1')
plt.show()

Polarity Values on Interactive map with geopandas .explore()

bitcoin_tweet2.explore(column='polarity',   # make choropleth based on "polarity" column
    tooltip=['polarity'],  # show "polarity" value in tooltip (on hover)
    popup=True,  # show all values in popup (on click)
    tiles="openstreetmap",  # use "openstreetmap" tiles
    cmap="Set1",  # use "Set1" matplotlib colormap
    legend=True,  
    marker_kwds=dict(radius=5,icon=folium.Icon(icon='house-blank')),                    
    style_kwds=dict(color="black"), # use black outline
    zoom_control=True,
    #zoom_start = 1,
                    )

Make this Notebook Trusted to load map: File -> Trust Notebook

Polarity Labels

fig, ax = plt.subplots(figsize=(20,16))
earth.plot(ax=ax, color= "grey")
bitcoin_tweet2.plot(ax=ax, column='polarity_label', legend=True,  cmap='viridis')
plt.show()

Polarity Values on Interactive Map with GeoPandas .explore()

bitcoin_tweet2.explore(column='polarity_label',   # make choropleth based on "polarity_label" column
    tooltip=['polarity_label'],  # show "polarity_label" value in tooltip (on hover)
    popup=True,  # show all values in popup (on click)
    tiles="openstreetmap",  # use "openstreetmap" tiles
    cmap="viridis",  # use "Set1" matplotlib colormap
    legend=True,  
    marker_kwds=dict(radius=5,icon=folium.Icon(icon='house-blank')),                    
    style_kwds=dict(color="black"), # use black outline
    zoom_control=True,
    #zoom_start = 1,
                      
                      )

Make this Notebook Trusted to load map: File -> Trust Notebook

Task 2.4 Subjectivity Analysis

Calculate the subjectivity values of all the tweets. For a given geographical location, if you have more than one tweet then find the average subjectivity value taking into consideration all the tweets generated from the same location. Using a suitable plot type (such as a geographical map), perform a geospatial visualisation of the subjectivities corresponding to all the tweets. Whilst you are free to choose a plot type, the visualisation must be clear and easy to understand/interpret.

Define a function that extracts subjectivity values from the tweets(text)

def getTextSubjectivity(txt):
    return TextBlob(txt).sentiment.subjectivity

Apply the defined function and inspect

bitcoin_tweet['subjectivity'] = bitcoin_tweet['text'].apply(getTextSubjectivity)
bitcoin_tweet

	field_1	user_location	text	user_followers	locations	latitude	longitude	geometry	polarity	polarity_label	subjectivity
0	714437	The Moon, Sylacauga, Talladega County, Alabama...	talked to about stacks stx and hopes the...	39	The Moon	33.1584497	-86.2385846	POINT (-86.23858 33.15845)	-0.125000	Negative	0.375000
1	280261	Lisboa, Portugal	i m so f tired of ignorant american phd econom...	1614	Lisbon, Portugal	38.7077507	-9.1365919	POINT (-9.13659 38.70775)	-0.133333	Negative	0.233333
2	336653	Muhu, Saare maakond, Eesti	long bitcoin	43	Moon	58.5959044	23.21964608602439	POINT (23.21965 58.59590)	-0.050000	Negative	0.400000
3	872349	Global, 81, Barber Greene Road, Don Mills, Don...	bitcoin bull cathie wood attracts big short...	77240	Global	43.7283874	-79.34914879325001	POINT (-79.34915 43.72839)	0.000000	Neutral	0.200000
4	109405	Canada	bitcoin price in us dollar btc usd btcusd ...	250	Canada	61.0666922	-107.991707	POINT (-107.99171 61.06669)	0.165000	Positive	0.351667
...	...	...	...	...	...	...	...	...	...	...	...
495	898076	Hyderabad, Bahadurpura mandal, Hyderabad Distr...	freedom 35 official bsc bitcoin via	112	Hyderabad, India	17.360589	78.4740613	POINT (78.47406 17.36059)	0.000000	Neutral	0.000000
496	951645	b'424c4f434b434841494e2c20d09dd0b0d0b1d0b5d180...	bitcoin btc i don t like these lower highs...	21254	#blockchain	44.6465984	34.4007341	POINT (34.40073 44.64660)	0.000000	Neutral	0.000000
497	7974	Sydney, Council of the City of Sydney, New Sou...	what goes up must comes down hard question ...	26	Sydney, New South Wales	-33.8698439	151.2082848	POINT (151.20828 -33.86984)	-0.223611	Negative	0.415278
498	788995	Montréal, Agglomération de Montréal, MontrÃ...	am i the only one that likes watching the sats...	135	Montréal, Québec	45.5031824	-73.5698065	POINT (-73.56981 45.50318)	0.250000	Positive	0.750000
499	62652	Laguna Beach, Orange County, California, Unite...	lamb is a fast safe and scalable blockchai...	3784	Laguna Beach, CA	33.5426975	-117.785366	POINT (-117.78537 33.54270)	0.350000	Positive	0.550000

500 rows × 11 columns

Define a function a that labels the subjective values, (Oluyale, 2023).

def  definesubjectivity(x):
  if x > 0.5:
    return "subjective" 
  elif x < 0.5: 
    return "Factual" 
  elif x == 0.5: 
    return "Neutral" 

bitcoin_tweet #inspect

	field_1	user_location	text	user_followers	locations	latitude	longitude	geometry	polarity	polarity_label	subjectivity
0	714437	The Moon, Sylacauga, Talladega County, Alabama...	talked to about stacks stx and hopes the...	39	The Moon	33.1584497	-86.2385846	POINT (-86.23858 33.15845)	-0.125000	Negative	0.375000
1	280261	Lisboa, Portugal	i m so f tired of ignorant american phd econom...	1614	Lisbon, Portugal	38.7077507	-9.1365919	POINT (-9.13659 38.70775)	-0.133333	Negative	0.233333
2	336653	Muhu, Saare maakond, Eesti	long bitcoin	43	Moon	58.5959044	23.21964608602439	POINT (23.21965 58.59590)	-0.050000	Negative	0.400000
3	872349	Global, 81, Barber Greene Road, Don Mills, Don...	bitcoin bull cathie wood attracts big short...	77240	Global	43.7283874	-79.34914879325001	POINT (-79.34915 43.72839)	0.000000	Neutral	0.200000
4	109405	Canada	bitcoin price in us dollar btc usd btcusd ...	250	Canada	61.0666922	-107.991707	POINT (-107.99171 61.06669)	0.165000	Positive	0.351667
...	...	...	...	...	...	...	...	...	...	...	...
495	898076	Hyderabad, Bahadurpura mandal, Hyderabad Distr...	freedom 35 official bsc bitcoin via	112	Hyderabad, India	17.360589	78.4740613	POINT (78.47406 17.36059)	0.000000	Neutral	0.000000
496	951645	b'424c4f434b434841494e2c20d09dd0b0d0b1d0b5d180...	bitcoin btc i don t like these lower highs...	21254	#blockchain	44.6465984	34.4007341	POINT (34.40073 44.64660)	0.000000	Neutral	0.000000
497	7974	Sydney, Council of the City of Sydney, New Sou...	what goes up must comes down hard question ...	26	Sydney, New South Wales	-33.8698439	151.2082848	POINT (151.20828 -33.86984)	-0.223611	Negative	0.415278
498	788995	Montréal, Agglomération de Montréal, MontrÃ...	am i the only one that likes watching the sats...	135	Montréal, Québec	45.5031824	-73.5698065	POINT (-73.56981 45.50318)	0.250000	Positive	0.750000
499	62652	Laguna Beach, Orange County, California, Unite...	lamb is a fast safe and scalable blockchai...	3784	Laguna Beach, CA	33.5426975	-117.785366	POINT (-117.78537 33.54270)	0.350000	Positive	0.550000

500 rows × 11 columns

Apply the defined function

bitcoin_tweet["subjectivity_label"] = bitcoin_tweet["subjectivity"].apply(definesubjectivity)

Plot a histogram showing subjectivity values count

sns.histplot(x='subjectivity', data= bitcoin_tweet, color='orange')

<Axes: xlabel='subjectivity', ylabel='Count'>

Do a count plot of subjectivity label

plt.figure(figsize=(12,4))
sns.countplot(x='subjectivity_label', data=bitcoin_tweet)
plt.title('Subjectivity Label Count')
plt.show()

bitcoin_tweet.columns #inspect available columns

Index(['field_1', 'user_location', 'text', 'user_followers', 'locations',
       'latitude', 'longitude', 'geometry', 'polarity', 'polarity_label',
       'subjectivity', 'subjectivity_label'],
      dtype='object')

Choose columns necessary for map visualization

bitcoin_tweet3= bitcoin_tweet[['geometry', 'subjectivity', 'subjectivity_label']]

Map Visualisation of Subjectivity

Plotting subjectivity values

fig, ax = plt.subplots(figsize=(20,16))
earth.plot(ax=ax, color= "grey")
bitcoin_tweet3.plot(ax=ax, column='subjectivity', legend=True, legend_kwds={"label": 'Subjectivity Spread of Bitcoin Tweets', "orientation":"horizontal"}, cmap='Set1')
plt.show()

Plotting subjectivity values with interactive map using geopandas .explore()

bitcoin_tweet3.explore(column='subjectivity',   # make choropleth based on "BoroName" column
    tooltip=['subjectivity'],  # show "BoroName" value in tooltip (on hover)
    popup=True,  # show all values in popup (on click)
    tiles="openstreetmap",  # use "CartoDB positron" tiles
    cmap="Set1",  # use "Set1" matplotlib colormap
    legend=True,  
    marker_kwds=dict(radius=5,icon=folium.Icon(icon='house-blank')),                    
    style_kwds=dict(color="black"), # use black outline
    zoom_control=True,
    #zoom_start = 1,
        )

Make this Notebook Trusted to load map: File -> Trust Notebook

Plotting subjectivity label

fig, ax = plt.subplots(figsize=(20,16))
earth.plot(ax=ax, color= "grey")
bitcoin_tweet3.plot(ax=ax, column='subjectivity_label', legend=True,  cmap='viridis')
plt.show()

Plotting subjectivity label with interactive map using geopandas .explore()

bitcoin_tweet3.explore(column='subjectivity_label',   # make choropleth based on "BoroName" column
    tooltip=['subjectivity_label'],  # show "BoroName" value in tooltip (on hover)
    popup=True,  # show all values in popup (on click)
    tiles="openstreetmap",  # use "CartoDB positron" tiles
    cmap="Set1",  # use "Set1" matplotlib colormap
    legend=True,  
    marker_kwds=dict(radius=5,icon=folium.Icon(icon='house-blank')),                    
    style_kwds=dict(color="black"), # use black outline
    zoom_control=True,
    #zoom_start = 1,
                      )

Make this Notebook Trusted to load map: File -> Trust Notebook

Reading a GeoDataFrame from a DataFrame with coordinates

A GeoDataFrame needs a shapely object. We use geopandas points_from_xy() to transform Longitude and Latitude into a list of shapely.Point objects and set it as a geometry while creating the GeoDataFrame. (note that points_from_xy() is an enhanced wrapper for [Point(x, y) for x, y in zip(df.Longitude, df.Latitude)]). The crs value is also set to explicitly state the geometry data defines latitude/ longitude world geodetic degree values. This is important for the correct interpretation of the data, such as when plotting with data in other formats.

Now we interpret the sentiment based on the compound score. If the compound score is greater than or equal to 0.05, it's considered positive. If it's less than or equal to -0.05, it's considered negative. Otherwise, it's considered neutral. (Oluyale, 2023)

Oluyale, D. (2023, September 16). Sentiment Analysis using various options in Python Machine Learning. Medium. https://medium.com/@oluyaled/sentiment-analysis-using-various-options-in-python-machine-learning-aaa24ea0991c

Task 2.5 Storify/Interpretation

In this task, use your geospatial data analytical skill to storify (in not more than 500 words) the results obtained in the preceding two tasks. Imagine yourself as a policy advisor to the UK government whose job is to update about the public sentiment related to cryptocurrency across different parts of the world. You may try to answer some of these example questions – How is the public opinion about cryptocurrency? Which locations have positive views about this issue and where can you see a vast amount of negativity? Despite having positive/negative/mixed sentiment about cryptocurrency, will you take these tweets very seriously (HINT: if the tweet originates from outside the UK, then it may not affect the government policies!)? Are the messages loud and clear? Please note that these are only suggestive questions. You are strongly recommended to not constrain your sentiment analytical skills

Storification task would require more codes and visualization

print(bitcoin_tweet['locations'].value_counts().to_markdown())

| locations                                           |   count |
|:----------------------------------------------------|--------:|
| Bay Area, CA                                        |      19 |
| London, England                                     |      15 |
| United States                                       |      13 |
| Global                                              |      10 |
| New York, NY                                        |      10 |
| Australia                                           |       8 |
| Matter Doesn't Matter                               |       6 |
| Moon                                                |       6 |
| New York                                            |       6 |
| Earth                                               |       6 |
| The Moon                                            |       5 |
| England, United Kingdom                             |       5 |
| United Kingdom                                      |       5 |
| Blockchain                                          |       5 |
| India                                               |       5 |
| Birmingham, England                                 |       4 |
| Worldwide                                           |       4 |
| Europe                                              |       4 |
| California, USA                                     |       4 |
| Florida, USA                                        |       4 |
| New York, USA                                       |       4 |
| Mars                                                |       4 |
| Trade here 👉                                     |       4 |
| Singapore                                           |       4 |
| UK                                                  |       4 |
| Pakistan                                            |       3 |
| Miami, FL                                           |       3 |
| Chicago, IL                                         |       3 |
| Paris, FR                                           |       3 |
| Canada                                              |       3 |
| Bitcoin                                             |       3 |
| London                                              |       3 |
| Islamabad, Pakistan                                 |       3 |
| USA                                                 |       3 |
| World                                               |       3 |
| Kolkata, India                                      |       2 |
| Sankt-Peterburg                                     |       2 |
| Türkiye                                            |       2 |
| New York City                                       |       2 |
| San Diego, CA                                       |       2 |
| London, UK                                          |       2 |
| To the Moon                                         |       2 |
| Bangladesh                                          |       2 |
| Internet                                            |       2 |
| Houston, TX                                         |       2 |
| Paris, France                                       |       2 |
| Tennessee                                           |       2 |
| Austin                                              |       2 |
| tesvikiye                                           |       2 |
| America                                             |       2 |
| Bellevue, WA                                        |       2 |
| Spain                                               |       2 |
| Buffalo, NY                                         |       2 |
| São Paulo, Brasil 🇧🇷                         |       2 |
| London, England 🇬🇧                            |       2 |
| Dhaka, Bangladesh                                   |       2 |
| Minnesota                                           |       2 |
| on an island 🇨🇦                               |       2 |
| Manhattan, NY                                       |       2 |
| Rotterdam, Nederland                                |       2 |
| Nova Scotia, Canada                                 |       2 |
| Everywhere                                          |       2 |
| Hong Kong                                           |       2 |
| South Africa                                        |       2 |
| Crypto World                                        |       2 |
| Johannesburg, South Africa                          |       2 |
| Jersey City, New Jersey                             |       2 |
| Sydney, New South Wales                             |       2 |
| Toronto, Ontario                                    |       2 |
| Estados Unidos                                      |       2 |
| Kansas City, MO                                     |       2 |
| somewhere                                           |       2 |
| Philadelphia, PA                                    |       2 |
| Metaverse                                           |       1 |
| 👇YouTube Channel👇                             |       1 |
| Paris                                               |       1 |
| Parts Unknown                                       |       1 |
| भारत                                        |       1 |
| 127.0.0.1                                           |       1 |
| Islamic Republic of Iran                            |       1 |
| Metz / Paris, FRANCE                                |       1 |
| United States Of American                           |       1 |
| El Salvador                                         |       1 |
| Penzance, Cornwall                                  |       1 |
| Western Europe                                      |       1 |
| east Godavari, Andhra Pradesh                       |       1 |
| New Delhi, India                                    |       1 |
| Cambridge, MA                                       |       1 |
| Galt’s Gulch                                      |       1 |
| Orange County, California                           |       1 |
| Near You                                            |       1 |
| Ukraine                                             |       1 |
| Milan, Lombardy                                     |       1 |
| Cuddapah, India                                     |       1 |
| Bali, Indonesia                                     |       1 |
| DE 🇩🇪 & SA🇿🇦                            |       1 |
| Santa Monica, CA                                    |       1 |
| SPORTS BETTING (THE CASINO)                         |       1 |
| Grindelwald, Schweiz                                |       1 |
| Jawa Tengah, Indonesia                              |       1 |
| Traveler..                                          |       1 |
| Texas, USA                                          |       1 |
| Raigarh, India                                      |       1 |
| New York, NC                                        |       1 |
| U.S.A!                                              |       1 |
| Northern California                                 |       1 |
| b'd0bad0b0d0b7d0b0d185d181d182d0b0d0bd'             |       1 |
| Lausanne (Switzerland)                              |       1 |
| Worldwide                                           |       1 |
| Nouvelle calédonie                                 |       1 |
| DKI Jakarta, Indonesia                              |       1 |
| Copenhagen                                          |       1 |
| Bandung, Jawa Barat                                 |       1 |
| Los Angeles                                         |       1 |
| Morocco                                             |       1 |
| Russia                                              |       1 |
| Wayne, PA                                           |       1 |
| Dublin, Ireland                                     |       1 |
| Loyalsock Township, PA                              |       1 |
| Chile                                               |       1 |
| EARTH                                               |       1 |
| Brisbane                                            |       1 |
| East Midlands, England                              |       1 |
| Brookline, MA                                       |       1 |
| Faridpur, Bangladesh                                |       1 |
| Ciudad Real, Spain                                  |       1 |
| Washington, DC                                      |       1 |
| INDONESIA                                           |       1 |
| Europa                                              |       1 |
| Massachusetts                                       |       1 |
| Toronto                                             |       1 |
| Uranus                                              |       1 |
| South Korea                                         |       1 |
| Down the Rabbit Hole                                |       1 |
| Moon, PA                                            |       1 |
| Glasgow, Scotland                                   |       1 |
| London, United Kingdom                              |       1 |
| Sukabumi, Indonesia                                 |       1 |
| Anaheim, CA                                         |       1 |
| The Netherlands                                     |       1 |
| Jaipur Rajasthan                                    |       1 |
| Seattle, WA                                         |       1 |
| Detroit Michigan                                    |       1 |
| Cincinnati, OH                                      |       1 |
| Space Mountain                                      |       1 |
| Melbourne, Victoria                                 |       1 |
| Telegram                                            |       1 |
| Road Warrior                                        |       1 |
| ¯\_(ツ)_/¯                                       |       1 |
| Germany                                             |       1 |
| Ann Arbor                                           |       1 |
| Everywhere 🗺                                     |       1 |
| Tamil Nadu                                          |       1 |
| Hyderabad, India                                    |       1 |
| #blockchain                                         |       1 |
| Montréal, Québec                                  |       1 |
| British Columbia, Canada                            |       1 |
| Florida                                             |       1 |
| Sumatera Selatan, Indonesia                         |       1 |
| Columbia, SC                                        |       1 |
| Little Rock, AR                                     |       1 |
| Istanbul, Turkey                                    |       1 |
| Tucson, AZ                                          |       1 |
| South Park, CO                                      |       1 |
| Los Angeles, CA                                     |       1 |
| 420 Wall St, NY                                     |       1 |
| Spratly Islands                                     |       1 |
| Malang, Jawa Timur                                  |       1 |
| Ondo, Nigeria                                       |       1 |
| Texas                                               |       1 |
| Curaçao                                            |       1 |
| Nigeria                                             |       1 |
| Oslo, Norway                                        |       1 |
| Irving, Tx                                          |       1 |
| World Wide                                          |       1 |
| earth                                               |       1 |
| Tegal                                               |       1 |
| Santa Fe, New Mexico                                |       1 |
| Sydney, Australia                                   |       1 |
| Bangkok, Thailand                                   |       1 |
| Jakarta, Indonesia                                  |       1 |
| Fairfax, VA                                         |       1 |
| Strazburg, France                                   |       1 |
| Los Angeles, California, USA                        |       1 |
| in search                                           |       1 |
| bella ciao                                          |       1 |
| Barcelona / Bangkok                                 |       1 |
| International Space Station 🚀                    |       1 |
| Larissa, GR                                         |       1 |
| South East, England                                 |       1 |
| Switzerland                                         |       1 |
| Maui, Hawaii                                        |       1 |
| Shanghai, China                                     |       1 |
| Milan, Italy                                        |       1 |
| Oklahoma City, OK                                   |       1 |
| Busan, Republic of Korea                            |       1 |
| West Dhanmondi, Dhaka                               |       1 |
| Bumi Nusantara                                      |       1 |
| Austin, TX                                          |       1 |
| Brussel, België                                    |       1 |
| Dallas, TX                                          |       1 |
| Kyiv, Ukraine                                       |       1 |
| Rangpur, Bangladesh                                 |       1 |
| Glasgow, UK                                         |       1 |
| Zug | Berlin                                        |       1 |
| Landgraaf, Nederland                                |       1 |
| here                                                |       1 |
| Argentina                                           |       1 |
| Washington D.C                                      |       1 |
| Shambhala                                           |       1 |
| Planet Earth                                        |       1 |
| American Fork, UT                                   |       1 |
| Salt Lake City, Utah                                |       1 |
| Tehran                                              |       1 |
| Garmany                                             |       1 |
| Utah                                                |       1 |
| Internationalist                                    |       1 |
| Moon                                                |       1 |
| 대한민국 안산시                              |       1 |
| Pekanbaru, Riau                                     |       1 |
| united states                                       |       1 |
| Alger                                               |       1 |
| Global                                              |       1 |
| Meta                                                |       1 |
| Mother Earth                                        |       1 |
| Punjab, Pakistan                                    |       1 |
| Bareilly, India                                     |       1 |
| @Moon                                               |       1 |
| Turkey                                              |       1 |
| Boston, MA                                          |       1 |
| California, USA🇺🇸                             |       1 |
| San Tan Valley, AZ                                  |       1 |
| الولايات المتحدة الأمريكية  |       1 |
| Samsun, Türkiye                                    |       1 |
| Future                                              |       1 |
| San Mateo, CA                                       |       1 |
| Westland, MI                                        |       1 |
| Miami Florida                                       |       1 |
| Mewn                                                |       1 |
| Guess                                               |       1 |
| Burger-galaxy                                       |       1 |
| Ireland                                             |       1 |
| Pace, FL                                            |       1 |
| Watford, Hertfordshire, UK                          |       1 |
| Malaysia                                            |       1 |
| Rio de Janeiro, Brazil                              |       1 |
| Lewes, DE                                           |       1 |
| Between Here & There                                |       1 |
| Sydney                                              |       1 |
| Bay Area, California                                |       1 |
| Mumbai                                              |       1 |
| Dubai, United Arab Emirates                         |       1 |
| Gazipur, Dhaka                                      |       1 |
| Kansas City                                         |       1 |
| Tucuman, Argentina                                  |       1 |
| Ohio, USA                                           |       1 |
| België                                             |       1 |
| Washington state                                    |       1 |
| Jackson,Mississippi,USA                             |       1 |
| Antwerp                                             |       1 |
| Englewood Cliffs, NJ                                |       1 |
| India                                               |       1 |
| West Bengal, India                                  |       1 |
| İstanbul, Türkiye                                 |       1 |
| Mumbai, India                                       |       1 |
| Ahmedabad City, India                               |       1 |
| #الكرة_الارضية                          |       1 |
| Amsterdam, Nederland                                |       1 |
| Wellington City, New Zealand                        |       1 |
| Ankara, Türkiye                                    |       1 |
| South West, England                                 |       1 |
| halp, USA                                           |       1 |
| Durham, NC                                          |       1 |
| Texas                                               |       1 |
| Greenville, SC                                      |       1 |
| india                                               |       1 |
| Wyoming                                             |       1 |
| AZ/NH/MA                                            |       1 |
| THE MOON                                            |       1 |
| Out and About                                       |       1 |
| Space                                               |       1 |
| City of London, London                              |       1 |
| Windsor, Ontario                                    |       1 |
| b'536f7574682041667269636120f09f87bff09f87a6e29da4' |       1 |
| Hamburg, Deutschland                                |       1 |
| Nova Friburgo, Brasil                               |       1 |
| Gas Fields, Louisiana                               |       1 |
| now                                                 |       1 |
| Lagos                                               |       1 |
| New Orleans                                         |       1 |
| Jaderberg                                           |       1 |
| Chattogram , Bangladesh                             |       1 |
| Michigan, USA                                       |       1 |
| 3rd rock from the sun                               |       1 |
| Islington, London                                   |       1 |
| Bogor                                               |       1 |
| Fort Worth, Texas                                   |       1 |
| New York & Taipei                                   |       1 |
| Venezuela                                           |       1 |
| Indiana, USA                                        |       1 |
| Sri lanka                                           |       1 |
| Lisbon, Portugal                                    |       1 |
| Laguna Beach, CA                                    |       1 |

bitcoin_tweet[:3]

	field_1	user_location	text	user_followers	locations	latitude	longitude	geometry	polarity	polarity_label	subjectivity	subjectivity_label
0	714437	The Moon, Sylacauga, Talladega County, Alabama...	talked to about stacks stx and hopes the...	39	The Moon	33.1584497	-86.2385846	POINT (-86.23858 33.15845)	-0.125000	Negative	0.375000	Factual
1	280261	Lisboa, Portugal	i m so f tired of ignorant american phd econom...	1614	Lisbon, Portugal	38.7077507	-9.1365919	POINT (-9.13659 38.70775)	-0.133333	Negative	0.233333	Factual
2	336653	Muhu, Saare maakond, Eesti	long bitcoin	43	Moon	58.5959044	23.21964608602439	POINT (23.21965 58.59590)	-0.050000	Negative	0.400000	Factual

bri= bitcoin_tweet[(bitcoin_tweet["locations"]== 'London, England')] 
                   

bri

	field_1	user_location	text	user_followers	locations	latitude	longitude	geometry	polarity	polarity_label	subjectivity	subjectivity_label
20	419791	London, Greater London, England, United Kingdom	thanks for giving us such a great opportunit...	129	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	3.100000e-01	Positive	0.340000	Factual
54	295173	London, Greater London, England, United Kingdom	bitcoin fear index on google search statistics...	202	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	3.750000e-01	Positive	0.500000	Neutral
103	820864	London, Greater London, England, United Kingdom	no trade btc price is 30888 at time 08 07 21...	1523	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	0.000000e+00	Neutral	0.000000	Factual
112	505680	London, Greater London, England, United Kingdom	grayscale pairs with coindesk index to launch ...	62	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	0.000000e+00	Neutral	0.000000	Factual
117	271016	London, Greater London, England, United Kingdom	it was good while it lasted now it is time ...	129	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	3.500000e-01	Positive	0.300000	Factual
234	513485	London, Greater London, England, United Kingdom	grayscale investments launches defi fund now...	62	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	1.000000e-01	Positive	0.000000	Factual
264	520206	London, Greater London, England, United Kingdom	ready for elon to say some stupid shit trigger...	63	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	-2.666667e-01	Negative	0.766667	subjective
322	824028	London, Greater London, England, United Kingdom	let the fifth and final wave commence btc bi...	7	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	0.000000e+00	Neutral	1.000000	subjective
362	568086	London, Greater London, England, United Kingdom	all details on telegram channel entry tar...	119	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	0.000000e+00	Neutral	0.000000	Factual
377	663473	London, Greater London, England, United Kingdom	the first ever mode post agm investor present...	788	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	3.250000e-01	Positive	0.366667	Factual
378	272275	London, Greater London, England, United Kingdom	open spot signals 10 ada rlc ...	78	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	4.000000e-01	Positive	0.450000	Factual
386	578083	London, Greater London, England, United Kingdom	the fiat money experiment is failing bitcoin	18	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	0.000000e+00	Neutral	0.000000	Factual
398	624101	London, Greater London, England, United Kingdom	huge bitcoin inflow to gemini behind the drop ...	8249	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	2.775558e-17	Positive	0.800000	subjective
436	558763	London, Greater London, England, United Kingdom	we don t buy bitcoin due to the new unique...	851	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	2.385281e-01	Positive	0.594697	subjective
466	130805	London, Greater London, England, United Kingdom	elsalvador becomes the first country to make ...	469	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	2.250000e-01	Positive	0.266667	Factual

bri2 = bitcoin_tweet[(bitcoin_tweet["locations"]== 'United Kingdom')]

bri2

	field_1	user_location	text	user_followers	locations	latitude	longitude	geometry	polarity	polarity_label	subjectivity	subjectivity_label
127	349546	United Kingdom	what would the government do if the youth of t...	261	United Kingdom	54.7023545	-3.2765753	POINT (-3.27658 54.70235)	0.0	Neutral	0.0	Factual
137	531787	United Kingdom	swedish man sentenced for gold backed cryptocu...	173	United Kingdom	54.7023545	-3.2765753	POINT (-3.27658 54.70235)	0.0	Neutral	0.0	Factual
376	498148	United Kingdom	privacy focused crypto is launching an inc...	5840	United Kingdom	54.7023545	-3.2765753	POINT (-3.27658 54.70235)	0.0	Neutral	0.0	Factual
411	556330	United Kingdom	it s all profit cryptocurrency made me a milli...	2332	United Kingdom	54.7023545	-3.2765753	POINT (-3.27658 54.70235)	0.0	Neutral	0.0	Factual
460	412787	United Kingdom	buy bitcoin and hold newhigh	12	United Kingdom	54.7023545	-3.2765753	POINT (-3.27658 54.70235)	0.0	Neutral	0.0	Factual

br3 = bitcoin_tweet[(bitcoin_tweet["locations"]== 'London')]

br3

	field_1	user_location	text	user_followers	locations	latitude	longitude	geometry	polarity	polarity_label	subjectivity	subjectivity_label
36	891204	London, Greater London, England, United Kingdom	good traders vs bad traders a bad trade on th...	1463	London	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	-0.217857	Negative	0.539286	subjective
256	157176	London, Greater London, England, United Kingdom	you still tweeting about bitcoin wait u...	43	London	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	0.142857	Positive	0.267857	Factual
475	367564	London, Greater London, England, United Kingdom	so after a very long wait tomorrow is the sta...	76	London	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	0.267500	Positive	0.710000	subjective

br4 = bitcoin_tweet[(bitcoin_tweet["locations"]== 'London, England 🇬🇧')]
br4

	field_1	user_location	text	user_followers	locations	latitude	longitude	geometry	polarity	polarity_label	subjectivity	subjectivity_label
76	663255	London, Greater London, England, United Kingdom	i m encouraged by this morning s senate bankin...	4285	London, England 🇬🇧	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	0.0	Neutral	0.0	Factual
433	438568	London, Greater London, England, United Kingdom	ta bitcoin trim losses why bulls need to ove...	4299	London, England 🇬🇧	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	0.5	Positive	0.5	Neutral

br5 = bitcoin_tweet[(bitcoin_tweet["locations"]== 'England, United Kingdom')]
br5

	field_1	user_location	text	user_followers	locations	latitude	longitude	geometry	polarity	polarity_label	subjectivity	subjectivity_label
123	890689	England, United Kingdom	live bitcoin price 46 731 an increase ...	39938	England, United Kingdom	52.5310214	-1.2649062	POINT (-1.26491 52.53102)	0.068182	Positive	0.283333	Factual
159	739840	England, United Kingdom	fair launching today 6pm utc do not miss...	320	England, United Kingdom	52.5310214	-1.2649062	POINT (-1.26491 52.53102)	0.291667	Positive	0.550000	subjective
196	228331	England, United Kingdom	bitcoin wow some people woke up thinking thi...	7	England, United Kingdom	52.5310214	-1.2649062	POINT (-1.26491 52.53102)	0.141667	Positive	0.741667	subjective
423	293677	England, United Kingdom	live bitcoin price 35 494 an increase ...	39946	England, United Kingdom	52.5310214	-1.2649062	POINT (-1.26491 52.53102)	0.068182	Positive	0.283333	Factual
487	423690	England, United Kingdom	moving up slowly inertia creeps moving up slo...	5	England, United Kingdom	52.5310214	-1.2649062	POINT (-1.26491 52.53102)	-0.300000	Negative	0.400000	Factual

Concatenate the dataframes on rows and call it bitcoin_tweetUK

bitcoin_tweetUK = pd.concat([bri, bri2, br3, br4, br5], axis=0)

bitcoin_tweetUK

	field_1	user_location	text	user_followers	locations	latitude	longitude	geometry	polarity	polarity_label	subjectivity	subjectivity_label
20	419791	London, Greater London, England, United Kingdom	thanks for giving us such a great opportunit...	129	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	3.100000e-01	Positive	0.340000	Factual
54	295173	London, Greater London, England, United Kingdom	bitcoin fear index on google search statistics...	202	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	3.750000e-01	Positive	0.500000	Neutral
103	820864	London, Greater London, England, United Kingdom	no trade btc price is 30888 at time 08 07 21...	1523	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	0.000000e+00	Neutral	0.000000	Factual
112	505680	London, Greater London, England, United Kingdom	grayscale pairs with coindesk index to launch ...	62	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	0.000000e+00	Neutral	0.000000	Factual
117	271016	London, Greater London, England, United Kingdom	it was good while it lasted now it is time ...	129	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	3.500000e-01	Positive	0.300000	Factual
234	513485	London, Greater London, England, United Kingdom	grayscale investments launches defi fund now...	62	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	1.000000e-01	Positive	0.000000	Factual
264	520206	London, Greater London, England, United Kingdom	ready for elon to say some stupid shit trigger...	63	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	-2.666667e-01	Negative	0.766667	subjective
322	824028	London, Greater London, England, United Kingdom	let the fifth and final wave commence btc bi...	7	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	0.000000e+00	Neutral	1.000000	subjective
362	568086	London, Greater London, England, United Kingdom	all details on telegram channel entry tar...	119	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	0.000000e+00	Neutral	0.000000	Factual
377	663473	London, Greater London, England, United Kingdom	the first ever mode post agm investor present...	788	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	3.250000e-01	Positive	0.366667	Factual
378	272275	London, Greater London, England, United Kingdom	open spot signals 10 ada rlc ...	78	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	4.000000e-01	Positive	0.450000	Factual
386	578083	London, Greater London, England, United Kingdom	the fiat money experiment is failing bitcoin	18	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	0.000000e+00	Neutral	0.000000	Factual
398	624101	London, Greater London, England, United Kingdom	huge bitcoin inflow to gemini behind the drop ...	8249	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	2.775558e-17	Positive	0.800000	subjective
436	558763	London, Greater London, England, United Kingdom	we don t buy bitcoin due to the new unique...	851	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	2.385281e-01	Positive	0.594697	subjective
466	130805	London, Greater London, England, United Kingdom	elsalvador becomes the first country to make ...	469	London, England	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	2.250000e-01	Positive	0.266667	Factual
127	349546	United Kingdom	what would the government do if the youth of t...	261	United Kingdom	54.7023545	-3.2765753	POINT (-3.27658 54.70235)	0.000000e+00	Neutral	0.000000	Factual
137	531787	United Kingdom	swedish man sentenced for gold backed cryptocu...	173	United Kingdom	54.7023545	-3.2765753	POINT (-3.27658 54.70235)	0.000000e+00	Neutral	0.000000	Factual
376	498148	United Kingdom	privacy focused crypto is launching an inc...	5840	United Kingdom	54.7023545	-3.2765753	POINT (-3.27658 54.70235)	0.000000e+00	Neutral	0.000000	Factual
411	556330	United Kingdom	it s all profit cryptocurrency made me a milli...	2332	United Kingdom	54.7023545	-3.2765753	POINT (-3.27658 54.70235)	0.000000e+00	Neutral	0.000000	Factual
460	412787	United Kingdom	buy bitcoin and hold newhigh	12	United Kingdom	54.7023545	-3.2765753	POINT (-3.27658 54.70235)	0.000000e+00	Neutral	0.000000	Factual
36	891204	London, Greater London, England, United Kingdom	good traders vs bad traders a bad trade on th...	1463	London	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	-2.178571e-01	Negative	0.539286	subjective
256	157176	London, Greater London, England, United Kingdom	you still tweeting about bitcoin wait u...	43	London	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	1.428571e-01	Positive	0.267857	Factual
475	367564	London, Greater London, England, United Kingdom	so after a very long wait tomorrow is the sta...	76	London	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	2.675000e-01	Positive	0.710000	subjective
76	663255	London, Greater London, England, United Kingdom	i m encouraged by this morning s senate bankin...	4285	London, England 🇬🇧	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	0.000000e+00	Neutral	0.000000	Factual
433	438568	London, Greater London, England, United Kingdom	ta bitcoin trim losses why bulls need to ove...	4299	London, England 🇬🇧	51.5074456	-0.1277653	POINT (-0.12777 51.50745)	5.000000e-01	Positive	0.500000	Neutral
123	890689	England, United Kingdom	live bitcoin price 46 731 an increase ...	39938	England, United Kingdom	52.5310214	-1.2649062	POINT (-1.26491 52.53102)	6.818182e-02	Positive	0.283333	Factual
159	739840	England, United Kingdom	fair launching today 6pm utc do not miss...	320	England, United Kingdom	52.5310214	-1.2649062	POINT (-1.26491 52.53102)	2.916667e-01	Positive	0.550000	subjective
196	228331	England, United Kingdom	bitcoin wow some people woke up thinking thi...	7	England, United Kingdom	52.5310214	-1.2649062	POINT (-1.26491 52.53102)	1.416667e-01	Positive	0.741667	subjective
423	293677	England, United Kingdom	live bitcoin price 35 494 an increase ...	39946	England, United Kingdom	52.5310214	-1.2649062	POINT (-1.26491 52.53102)	6.818182e-02	Positive	0.283333	Factual
487	423690	England, United Kingdom	moving up slowly inertia creeps moving up slo...	5	England, United Kingdom	52.5310214	-1.2649062	POINT (-1.26491 52.53102)	-3.000000e-01	Negative	0.400000	Factual

bitcoin_tweetUK.shape

(30, 12)

I will get the polarity Label Count with respect to UK

plt.figure(figsize=(12,4))
sns.countplot(x='polarity_label', data=bitcoin_tweetUK)
plt.title('Polarity Label Count for UK')
plt.show()

Subjectivity Count for UK

plt.figure(figsize=(12,4))
sns.countplot(x='subjectivity_label', data=bitcoin_tweetUK)
plt.title('Subjectivity Label Count for UK')
plt.show()

Storify/Interpretation

Introduction

Bitcoin is part of a peer-to-peer network called cryptocurrency. Users can exchange value digitally without the intervention of a third party by using Bitcoin. The concept of Bitcoin relies on the theory of resolving cryptographic algorithms to produce distinct hashes with a limited quantity. This work is set out to review sentiments on Bitcoin.

My Review Findings:

1. The result of my analysis shows that there is a general awareness of Bitcoin all over the world.

2. The world is favourably disposed to Bitcoin; only about 60 tweets of the sampled 500 tweets expressed negative sentiment about Bitcoin.

3. About 230 tweets expressed positive sentiments, and 210 tweets were neutral.

4. I have also attempted to understand the degree of negativity expressed; about 80% expressed very low negativity, e.g., between -0.1 and -0.09.

5. The sample from the UK also showed that only about 2 tweets of the sampled 30 tweets expressed negative sentiments about Bitcoin. The degree of negativity was again low.

6. 16 tweets expressed positive sentiments, and 12 tweets were neutral.

7. I attempted to understand if these sentiments were factual or subjective.

8. The tweets were overwhelmingly factual; only about 130 tweets were subjective worldwide.

9. In the United Kingdom, only nine tweets were subjective.

10. Given the fact that the overwhelming tweets were factual in the UK and all over the world, we can 100% take this analysis seriously.

11. One negative tweet in the UK came from the Brighton area around Bexhill-on-Sea. The other negative tweet came from Leicester, an area around Hinckley. The positive tweets are evenly spread all over the UK.

12. Among the neighbouring countries to the UK, the Republic of Ireland does not have a negative tweet about Bitcoin. France and the Netherlands have one negative comment each.

Conclusion and Policy Implications

I have been able to present the sentiment polarity and subjectivity from the United Kingdom and all over the world on Bitcoin. I have shown that Bitcoin tweets in the UK and around the world are overwhelmingly positively polarised. The sentiments expressed are overwhelmingly factual. It therefore means that these tweets can be trusted and could form the bedrock of policy implementation. I hereby suggest that:

1. The UK government could sponsor research on the benefits of Bitcoin and cryptocurrencies to the growth of its economy.

2. Bitcoin could be regarded as a digital legal tender.

3. Universities in the UK should be encouraged to take Bitcoin, crypto-currency, and cryptography as a course of study.

4. Scholarships and grants should be given to students who wish to study bitcoin and its mechanism of cryptography.

5. Research should be done to make Bitcoin a more secure digital currency. The negative sentiments about Bitcoin were around the fear of its security.

6. There is also a fear of the fact that there is no central ownership of Bitcoin. I will advise the government to set up control bodies for Bitcoin to outweigh these fears.

References

Crickard, P., (2018). Mastering geospatial analysis with Python: Explore GIS processing and learn to work with GeoDjango, CARTOframes and MapboxGL Jupyter. Packt Publishing. Link Gabby, A. (2023). Python Sentiment Analysis using TextBlob and VADER for Glassdoor Reviews. Medium. https://medium.com/@gabya06/python-sentiment-analysis-using-textblob-and-vader-for-glassdoor-reviews-cc9632babb73

GeoPandas Documentation (2023). Geopandas.GeoDataFrame.explore — GeoPandas Docs. Available at: https://geopandas.org/en/stable/docs/reference/api/geopandas.GeoDataFrame.explore.html.

GeoPandas Documentation (2023). Creating a GeoDataFrame from a DataFrame with coordinates— GeoPandas Docs. Available at: Creating a GeoDataFrame from a DataFrame with coordinates — GeoPandas 0+untagged.50.g9a9f097.dirty documentation

Deparkes (2016). Folium Map Tiles - deparkes. Available at: https://deparkes.co.uk/2016/06/10/folium-map-tiles/.

Intelligent Economist, (2020). Malthusian Theory Of Population - Intelligent Economist. Available at: https://www.intelligenteconomist.com/malthusian-theory/

Muhammad, U. S. (2023). A Comparison of NLTK and TextBlob for Text Analysis. Medium. https://medium.com/@umarsmuhammed/a-comparison-of-nltk-and-textblob-for-text-analysis-bd9ebcd0ecd9

Oluyale, D. (2023, September 16). Sentiment Analysis using various options in Python Machine Learning. Medium. https://medium.com/@oluyaled/sentiment-analysis-using-various-options-in-python-machine-learning-aaa24ea0991c

re — Regular expression operations. (2023). Python Documentation. Aavailable at: https://docs.python.org/3/library/re.html Accessed: 29/11/2023 TextBlob: Simplified Text Processing — TextBlob 0.16.0 documentation. (2023). TextBlob: Simplified Text Processing — TextBlob 0.16.0 Documentation. https://textblob.readthedocs.io/en/dev/