Neural network based identification of terrorist groups using explainable artificial intelligence

Research output: Chapter in Book/Report/Conference proceedingConference contribution

73 Downloads (Pure)

Abstract

Currently machine learning (ML) and artificial intelligence (AI) are applied in diverse domains and governments believe these technologies can be applied in identifying terrorist groups. This paper reviews literature covering the varied use of ML algorithms and proposes a novel approach with a first-time application of a deep neural network (DNN) to an existing case study for the identification of terrorist groups. The research will further seek to explain how the neural network arrived at its decision using SHapley Additive exPlanations (SHAP). The results reveal DNN outperformed two benchmark models in terms of accuracy. SHAP was able to explain features that influenced the predicted results.

Original languageEnglish
Title of host publicationProceedings of the 2023 IEEE Conference on Artificial Intelligence
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages191-192
Number of pages2
ISBN (Electronic)9798350339840
ISBN (Print)9798350339857
DOIs
Publication statusPublished - 2 Aug 2023
Event2023 IEEE Conference on Artificial Intelligence, CAI 2023 - Santa Clara, United States
Duration: 5 Jun 20236 Jun 2023

Conference

Conference2023 IEEE Conference on Artificial Intelligence, CAI 2023
Country/TerritoryUnited States
CitySanta Clara
Period5/06/236/06/23

Keywords

  • Big Data
  • Deep Neural Network
  • Explainable AI (XAI)
  • Global Terrorism Database
  • Machine Learning
  • Terrorism Identification

Cite this