jhhuang96 / predicting-terrorist-attacks

Predicting terrorist attacks using the Global Terrorism Database (GTD).

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Predicting Terrorist Attacks

By: Thomas Skowronek

Program: M.S. Data Science - Regis University

Course: MSDS-696 Data Science Practicum II

Project Goal

The M.S. Data Science program relies on the R language for course work. However, Python is more relevant in my current professional work. My goal for the MSDS-696 Practicum II course is to learn the Python language, and implement an end-to-end data science project.

Project Overview

To facilitate this objective, I used the Global Terrorism Database (GTD), which is maintained by the National Consortium for the Study of Terrorism and Responses to Terrorism (START) at the University of Maryland. This is a large dataset that provides an opportunity to use Python for data cleansing, exploratory data analysis, visualizations, feature engineering and machine learning. In addition, selecting the topic of terrorism is applicable to the current state of events across the world.

Data Type and Characteristics

The data distribution from START contains 170,350 observations and 135 attributes describing terrorist attacks across the world. The data covers the time period between 1970 through 2016, except for 1993. The attributes consist of numeric, categorical, character, and time series values. A brief visual analysis shows many attributes contain missing values and some have a high cardinality.

Source Code

The complete source for this project is available below. The source data for the project is not included. However, it is available for download from the Global Terrorism Database website.

Presentation

About

Predicting terrorist attacks using the Global Terrorism Database (GTD).


Languages

Language:Jupyter Notebook 100.0%