Mar 29, 2022

Mapping Global Terrorism with Folium and Publicly Available Data

Overview

We spend a lot of time working with different types of data sets and creating compelling visuals, which is always better than presenting a gigantic spreadsheet. This blog focuses on a short script that can be used to map your data and present it interactively to your customer or teammates. While the use case here is Global Terrorism, this approach can be applied to any dataset with coordinates. Even if your data does not have coordinates, these can be created with addresses and or location names using GeoPandas.

Note: We do not cover GeoPandas in this blog. Perhaps next time.

Find the dataset

The dataset can be found publicly available on Kaggle. The link is listed below. You will need this dataset downloaded before you begin.

https://www.kaggle.com/datasets/START-UMD/gtd

Download the dataset

The download link is on the “Data” Tab.

Install the libraries

Getting started is easy if you have a python interpreter already installed. I prefer PyCharm by JetBrains®.

Simply install the following libraries using pip in your terminal:

“pip install pandas folium”

This will install two of the necessary libraries for this project. First is the industry-standard, industrial, strength data analysis and manipulation tool for data scientists and machine learning, called Pandas. You can read more about it here (https://pandas.pydata.org ). The other package is Folium, this tool brings together the power of Python and Leaflet maps. (https://python-visualization.github.io/folium/ ). The documentation provides examples and there are many more on the Internet.

Build the script

The code is relatively straightforward and commented below. If you'd like to download the script, you can find that on our Github.

Github

Import the necessary python libraries. If they are not installed, run "pip install pandas folium" in your terminal.

import pandas as pd
import folium
from folium.plugins import MarkerCluster

After downloading the dataset you will need to load it into a Pandas dataframe. The "encoding=" argument is important in this case so ensure you are using "ISO-8859-1"
We also drop any rows that are missing values, hence the ".dropna(thresh=2)". This means we want to drop any rows from the dataframe that have two NaN or Null values.

df = pd.read_csv('globalterrorismdb_0617dist_terms.csv',encoding='ISO-8859-1').dropna(thresh=2)

Create and instance of the Folium Map using the "cartodpositron" argument for tiles. This the base map we will be using for this project. There are others available. Reference: https://carto.com/

world_map = folium.Map(tiles="cartodbpositron")

Since we will be adding marker clusters to the map that show where the terrorism occurred, we will need to create an instance of this object as well.

marker_cluster = MarkerCluster().add_to(world_map)
Check your results

Since our dataset has "latitude" and "longitude" coordinates we will use those to create our markers.
The "for loop" below follows this logic. For the entire length of the dataframe overall, which gives us our range count. Set the "lat" an "long" variables to the value of each latitude and longitude value in the dataframe.
We set the radius of the marker and the popup_test to show the Country Name and the Terror Group. These values are also in the dataframe under the columns of "country_text" and "gname".
Finally we create an iFrame to show the popup text, with the min and max values and then we create each marker and add it to the map. This may take a long time and the output file will be very large (up to 200MB).

for i in range(len(df)):
    if pd.notna(df.iloc[i]['latitude']):
            lat = df.iloc[i]['latitude']
            long = df.iloc[i]['longitude']
            radius = 5
            popup_text = """Country : {}<br>
                    Terror_Group : {}<br>"""
            popup_text = popup_text.format(df.iloc[i]['country_txt'],df.iloc[i]['gname'])
            iframe = folium.IFrame(popup_text)
            popup = folium.Popup(iframe,min_width=200,max_width=200)
            folium.CircleMarker(location=[lat, long], radius=radius, popup=popup, fill=True).add_to(marker_cluster)

Finally we save the output of the map with the markers to the file name listed below:

world_map.save(outfile='GTD_Global_Terrorism_Folium.html')

Check your results

After the script is completed, you will have an interactive map with all the locations marked and clustered by their relative geography. You map should look like the image below. You can zoom in or click the cluster objects to narrow down the clusters by region. The accuracy should be close enough to street level, assuming the exact coordinates provided are accurate.