Neural network based prediction of terrorist attacks using explainable artificial intelligence

Anna Rosner; Alexander Gegov; Djamila Ouelhadj; Adrian Alan Hopgood; Serge Da Deppo

doi:10.1109/CAI54212.2023.00089

Neural network based prediction of terrorist attacks using explainable artificial intelligence

NATO Allied Command Transformation

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

86 Downloads (Pure)

Abstract

AI has transformed the field of terrorism prediction, allowing law enforcement agencies to identify potential threats much more quickly and accurately. This paper proposes a first-time application of a neural network to predict the “success” of a terrorist attack. The neural network attains an accuracy of 91.66% and an F1 score of 0.954. This accuracy and F1 score are higher than those achieved with alternative benchmark models. However, using AI for predictions in high- stakes decisions also has limitations, including possible biases and ethical concerns. Therefore, the explainable AI (XAI) tool LIME is used to provide more insights into the algorithm's inner workings.

Original language	English
Title of host publication	Proceedings of IEEE CAI 2023: Conference on Artificial Intelligence
Publisher	Institute of Electrical and Electronics Engineers Inc.
Number of pages	2
ISBN (Electronic)	9798350339840
ISBN (Print)	9798350339857
DOIs	https://doi.org/10.1109/CAI54212.2023.00089
Publication status	Published - 2 Aug 2023
Event	IEEE CAI2023: Conference on Artificial Intelligence - Santa Clara, United States Duration: 5 Jun 2023 → 6 Jun 2023 https://cai.ieee.org/2023/

Conference

Conference	IEEE CAI2023: Conference on Artificial Intelligence
Country/Territory	United States
City	Santa Clara
Period	5/06/23 → 6/06/23
Internet address	https://cai.ieee.org/2023/

Keywords

explainable AI
terrorism prediction
Global Terrorism Database (GTD)
LIME
neural networks

Access to Document

10.1109/CAI54212.2023.00089

IEEECAI2023_Rosner_et_al_AAM
© 2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Accepted author manuscript (Post-print), 268 KB

https://cai.ieee.org/2023/

Cite this

@inproceedings{59bfaf6105404ed9834e0a12a768bf07,

title = "Neural network based prediction of terrorist attacks using explainable artificial intelligence",

abstract = "AI has transformed the field of terrorism prediction, allowing law enforcement agencies to identify potential threats much more quickly and accurately. This paper proposes a first-time application of a neural network to predict the “success” of a terrorist attack. The neural network attains an accuracy of 91.66\% and an F1 score of 0.954. This accuracy and F1 score are higher than those achieved with alternative benchmark models. However, using AI for predictions in high- stakes decisions also has limitations, including possible biases and ethical concerns. Therefore, the explainable AI (XAI) tool LIME is used to provide more insights into the algorithm's inner workings.",

keywords = "explainable AI, terrorism prediction, Global Terrorism Database (GTD), LIME, neural networks",

author = "Anna Rosner and Alexander Gegov and Djamila Ouelhadj and Hopgood, \{Adrian Alan\} and \{Da Deppo\}, Serge",

year = "2023",

month = aug,

day = "2",

doi = "10.1109/CAI54212.2023.00089",

language = "English",

isbn = "9798350339857",

booktitle = "Proceedings of IEEE CAI 2023: Conference on Artificial Intelligence",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

address = "United States",

note = "IEEE CAI2023: Conference on Artificial Intelligence ; Conference date: 05-06-2023 Through 06-06-2023",

url = "https://cai.ieee.org/2023/",

}

Rosner, A , Gegov, A , Ouelhadj, D , Hopgood, AA & Da Deppo, S 2023, Neural network based prediction of terrorist attacks using explainable artificial intelligence. in Proceedings of IEEE CAI 2023: Conference on Artificial Intelligence. Institute of Electrical and Electronics Engineers Inc., IEEE CAI2023: Conference on Artificial Intelligence, Santa Clara, California, United States, 5/06/23. https://doi.org/10.1109/CAI54212.2023.00089

Neural network based prediction of terrorist attacks using explainable artificial intelligence. / Rosner, Anna ; Gegov, Alexander ; Ouelhadj, Djamila et al.
Proceedings of IEEE CAI 2023: Conference on Artificial Intelligence. Institute of Electrical and Electronics Engineers Inc., 2023.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Neural network based prediction of terrorist attacks using explainable artificial intelligence

AU - Rosner, Anna

AU - Gegov, Alexander

AU - Ouelhadj, Djamila

AU - Hopgood, Adrian Alan

AU - Da Deppo, Serge

PY - 2023/8/2

Y1 - 2023/8/2

N2 - AI has transformed the field of terrorism prediction, allowing law enforcement agencies to identify potential threats much more quickly and accurately. This paper proposes a first-time application of a neural network to predict the “success” of a terrorist attack. The neural network attains an accuracy of 91.66% and an F1 score of 0.954. This accuracy and F1 score are higher than those achieved with alternative benchmark models. However, using AI for predictions in high- stakes decisions also has limitations, including possible biases and ethical concerns. Therefore, the explainable AI (XAI) tool LIME is used to provide more insights into the algorithm's inner workings.

AB - AI has transformed the field of terrorism prediction, allowing law enforcement agencies to identify potential threats much more quickly and accurately. This paper proposes a first-time application of a neural network to predict the “success” of a terrorist attack. The neural network attains an accuracy of 91.66% and an F1 score of 0.954. This accuracy and F1 score are higher than those achieved with alternative benchmark models. However, using AI for predictions in high- stakes decisions also has limitations, including possible biases and ethical concerns. Therefore, the explainable AI (XAI) tool LIME is used to provide more insights into the algorithm's inner workings.

KW - explainable AI

KW - terrorism prediction

KW - Global Terrorism Database (GTD)

KW - LIME

KW - neural networks

U2 - 10.1109/CAI54212.2023.00089

DO - 10.1109/CAI54212.2023.00089

M3 - Conference contribution

SN - 9798350339857

BT - Proceedings of IEEE CAI 2023: Conference on Artificial Intelligence

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - IEEE CAI2023: Conference on Artificial Intelligence

Y2 - 5 June 2023 through 6 June 2023

ER -

Neural network based prediction of terrorist attacks using explainable artificial intelligence

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this