1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
|
#+title: Data Visualization: World Choropleth Map of Happiness
#+date: 2020-09-25
#+description: Exploring and visualizing data with Python.
#+filetags: :data:
* Background Information
The dataset (obtained from [[https://www.kaggle.com/unsdsn/world-happiness][Kaggle]]) used in this article contains a list of
countries around the world, their happiness rankings and scores, as well as
other national scoring measures.
Fields include:
- Overall rank
- Country or region
- GDP per capita
- Social support
- Healthy life expectancy
- Freedom to make life choices
- Generosity
- Perceptions of corruption
There are 156 records. Since there are ~195 countries in the world, we can see
that around 40 countries will be missing from this dataset.
* Install Packages
As always, run the =install= command for all packages needed to perform
analysis.
#+begin_src python
!pip install folium geopandas matplotlib numpy pandas
#+end_src
* Import the Data
We only need a couple packages to create a choropleth map. We will use [[https://python-visualization.github.io/folium/][Folium]],
which provides map visualizations in Python. We will also use geopandas and
pandas to wrangle our data before we put it on a map.
#+begin_src python
# Import the necessary Python packages
import folium
import geopandas as gpd
import pandas as pd
#+end_src
To get anything to show up on a map, we need a file that will specify the
boundaries of each country. Luckily, GeoJSON files exist (for free!) on the
internet. To get the boundaries of every country in the world, we will use the
GeoJSON link shown below.
GeoPandas will take this data and load it into a dataframe so that we can easily
match it to the data we're trying to analyze. Let's look at the GeoJSON
dataframe:
#+begin_src python
# Load the GeoJSON data with geopandas
geo_data = gpd.read_file('https://raw.githubusercontent.com/datasets/geo-countries/master/data/countries.geojson')
geo_data.head()
#+end_src
#+caption: GeoJSON Dataframe
[[https://img.cleberg.net/blog/20200925-world-choropleth-map/geojson_df.png]]
Next, let's load the data from the Kaggle dataset. I've downloaded this file, so
update the file path if you have it somewhere else. After loading, let's take a
look at this dataframe:
#+begin_src python
# Load the world happiness data with pandas
happy_data = pd.read_csv(r'~/Downloads/world_happiness_data_2019.csv')
happy_data.head()
#+end_src
#+caption: Happiness Dataframe
[[https://img.cleberg.net/blog/20200925-world-choropleth-map/happiness_df.png]]
* Clean the Data
Some countries need to be renamed, or they will be lost when you merge the
happiness and GeoJSON dataframes. This is something I discovered when the map
below showed empty countries. I searched both data frames for the missing
countries to see the naming differences. Any countries that do not have records
in the =happy_data= df will not show up on the map.
#+begin_src python
# Rename some countries to match our GeoJSON data
# Rename USA
usa_index = happy_data.index[happy_data['Country or region'] == 'United States']
happy_data.at[usa_index, 'Country or region'] = 'United States of America'
# Rename Tanzania
tanzania_index = happy_data.index[happy_data['Country or region'] == 'Tanzania']
happy_data.at[tanzania_index, 'Country or region'] = 'United Republic of Tanzania'
# Rename the Congo
republic_congo_index = happy_data.index[happy_data['Country or region'] == 'Congo (Brazzaville)']
happy_data.at[republic_congo_index, 'Country or region'] = 'Republic of Congo'
# Rename the DRC
democratic_congo_index = happy_data.index[happy_data['Country or region'] == 'Congo (Kinshasa)']
happy_data.at[democratic_congo_index, 'Country or region'] = 'Democratic Republic of the Congo'
#+end_src
* Merge the Data
Now that we have clean data, we need to merge the GeoJSON data with the
happiness data. Since we've stored them both in dataframes, we just need to call
the =.merge()= function.
We will also rename a couple columns, just so that they're a little easier to
use when we create the map.
#+begin_src python
# Merge the two previous dataframes into a single geopandas dataframe
merged_df = geo_data.merge(happy_data,left_on='ADMIN', right_on='Country or region')
# Rename columns for ease of use
merged_df = merged_df.rename(columns = {'ADMIN':'GeoJSON_Country'})
merged_df = merged_df.rename(columns = {'Country or region':'Country'})
#+end_src
#+caption: Merged Dataframe
[[https://img.cleberg.net/blog/20200925-world-choropleth-map/merged_df.png]]
* Create the Map
The data is finally ready to be added to a map. The code below shows the
simplest way to find the center of the map and create a Folium map object. The
important part is to remember to reference the merged dataframe for our GeoJSON
data and value data. The columns specify which geo data and value data to use.
#+begin_src python
# Assign centroids to map
x_map = merged_df.centroid.x.mean()
y_map = merged_df.centroid.y.mean()
print(x_map,y_map)
# Creating a map object
world_map = folium.Map(location=[y_map, x_map], zoom_start=2,tiles=None)
folium.TileLayer('CartoDB positron',name='Dark Map',control=False).add_to(world_map)
# Creating choropleth map
folium.Choropleth(
geo_data=merged_df,
name='Choropleth',
data=merged_df,
columns=['Country','Overall rank'],
key_on='feature.properties.Country',
fill_color='YlOrRd',
fill_opacity=0.6,
line_opacity=0.8,
legend_name='Overall happiness rank',
smooth_factor=0,
highlight=True
).add_to(world_map)
#+end_src
Let's look at the resulting map.
#+caption: Choropleth Map
[[https://img.cleberg.net/blog/20200925-world-choropleth-map/map.png]]
* Create a Tooltip on Hover
Now that we have a map set up, we could stop. However, I want to add a tooltip
so that I can see more information about each country. The =tooltip_data= code
below will show a popup on hover with all the data fields shown.
#+begin_src python
# Adding labels to map
style_function = lambda x: {'fillColor': '#ffffff',
'color':'#000000',
'fillOpacity': 0.1,
'weight': 0.1}
tooltip_data = folium.features.GeoJson(
merged_df,
style_function=style_function,
control=False,
tooltip=folium.features.GeoJsonTooltip(
fields=['Country'
,'Overall rank'
,'Score'
,'GDP per capita'
,'Social support'
,'Healthy life expectancy'
,'Freedom to make life choices'
,'Generosity'
,'Perceptions of corruption'
],
aliases=['Country: '
,'Happiness rank: '
,'Happiness score: '
,'GDP per capita: '
,'Social support: '
,'Healthy life expectancy: '
,'Freedom to make life choices: '
,'Generosity: '
,'Perceptions of corruption: '
],
style=('background-color: white; color: #333333; font-family: arial; font-size: 12px; padding: 10px;')
)
)
world_map.add_child(tooltip_data)
world_map.keep_in_front(tooltip_data)
folium.LayerControl().add_to(world_map)
# Display the map
world_map
#+end_src
The final image below will show you what the tooltip looks like whenever you
hover over a country.
#+caption: Choropleth Map Tooltip
[[https://img.cleberg.net/blog/20200925-world-choropleth-map/tooltip_map.png]]
|
