Applying Machine Learning to Predict and Analyze Terrorist Attacks
Author(s)
So, Martin
D'Mello, Collin
Cherry, Ryan
Advisor(s)
Editor(s)
Collections
Supplementary to:
Permanent Link
Abstract
The purpose of this project is to explore new methods to prevent terrorist attacks using machine learning. A dataset which contains terrorist attacks from 1970 to 2021 was used. We proposed three machine learning models to help military or intelligence agencies develop contingency plans: K-Means clustering, SVM and Logistic Regression, and Random Forest. From the results, we found that K-Means clustering can help key stakeholders identify common weapons used for different types of terrorist attacks. We also found that SVM is able to classify terrorist attacks at an accuracy rate of 80% and random forest being able to classify at 83.53%. In conclusion, our paper only highlights benefits of key stakeholders to apply for machine learning models. As threats in the 21st century become complex due to technology, more research is needed to develop a comprehensive plan to counter a rapidly-changing hostile environment.
Sponsor
Date
2024-12
Extent
Resource Type
Text
Resource Subtype
Paper
Research Report
Research Report
Rights Statement
Unless otherwise noted, all materials are protected under U.S. Copyright Law and all rights are reserved