Machine learning-based classification of extremism using explainable artificial intelligence

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

doi:10.1109/IS61756.2024.10705241

Machine learning-based classification of extremism using explainable artificial intelligence

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

Abstract

This paper presents the first published application of multiple existing machine learning methods to a subset of features taken from the Profiles of Individual Radicalization in the United States (PIRUS) database to predict the feature ‘violent’. The best- performing model in terms of accuracy is the Hist Gradient Boosting model, with an accuracy of 89.06%, which is an improvement of more than 2.5% compared to the benchmark application. Permutation Feature Importance (PFI) and the explanation framework SHAP were then applied to explain the model predictions. Using both of these techniques together allows for a holistic view of both the model’s inner workings and the impact of the features on the results.

Original language	English
Title of host publication	2024 IEEE 12th International Conference on Intelligent Systems (IS)
Editors	Vassil Sgurev, Vladimir Jotsov, Vincenzo Piuri, Luybka Doukovska, Radoslav Yoshinov
Publisher	Institute of Electrical and Electronics Engineers Inc.
Number of pages	7
ISBN (Electronic)	9798350350982
ISBN (Print)	9798350350999
DOIs	https://doi.org/10.1109/IS61756.2024.10705241
Publication status	Published - 9 Oct 2024
Event	12th IEEE International Conference on Intelligent Systems, IS 2024 - Varna, Bulgaria Duration: 29 Aug 2024 → 31 Aug 2024

Publication series

Name	International Conference on Intelligent Systems
Publisher	IEEE
ISSN (Print)	2832-4145
ISSN (Electronic)	2767-9802

Conference

Conference	12th IEEE International Conference on Intelligent Systems, IS 2024
Country/Territory	Bulgaria
City	Varna
Period	29/08/24 → 31/08/24

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 16 Peace, Justice and Strong Institutions

Keywords

Machine Learning
Extremism
PIRUS database
eXplainable AI
SHAP
Permutation Feature Importance

Access to Document

10.1109/IS61756.2024.10705241

Cite this

Rosner, A., Gegov, A., Hopgood, A. A., Lamptey, O., Ouelhadj, D., & Da Deppo, S. (2024). Machine learning-based classification of extremism using explainable artificial intelligence. In V. Sgurev, V. Jotsov, V. Piuri, L. Doukovska, & R. Yoshinov (Eds.), 2024 IEEE 12th International Conference on Intelligent Systems (IS) (International Conference on Intelligent Systems). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IS61756.2024.10705241

Rosner, Anna ; Gegov, Alexander ; Hopgood, Adrian Alan et al. / Machine learning-based classification of extremism using explainable artificial intelligence. 2024 IEEE 12th International Conference on Intelligent Systems (IS). editor / Vassil Sgurev ; Vladimir Jotsov ; Vincenzo Piuri ; Luybka Doukovska ; Radoslav Yoshinov. Institute of Electrical and Electronics Engineers Inc., 2024. (International Conference on Intelligent Systems).

@inproceedings{18e3f2e20b244df4996bcaee29cc9dec,

title = "Machine learning-based classification of extremism using explainable artificial intelligence",

abstract = "This paper presents the first published application of multiple existing machine learning methods to a subset of features taken from the Profiles of Individual Radicalization in the United States (PIRUS) database to predict the feature {\textquoteleft}violent{\textquoteright}. The best- performing model in terms of accuracy is the Hist Gradient Boosting model, with an accuracy of 89.06\%, which is an improvement of more than 2.5\% compared to the benchmark application. Permutation Feature Importance (PFI) and the explanation framework SHAP were then applied to explain the model predictions. Using both of these techniques together allows for a holistic view of both the model{\textquoteright}s inner workings and the impact of the features on the results.",

keywords = "Machine Learning, Extremism, PIRUS database, eXplainable AI, SHAP, Permutation Feature Importance",

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

note = "Publisher Copyright: {\textcopyright} 2024 IEEE.; 12th IEEE International Conference on Intelligent Systems, IS 2024 ; Conference date: 29-08-2024 Through 31-08-2024",

year = "2024",

month = oct,

day = "9",

doi = "10.1109/IS61756.2024.10705241",

language = "English",

isbn = "9798350350999",

series = "International Conference on Intelligent Systems",

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

editor = "Vassil Sgurev and Vladimir Jotsov and Vincenzo Piuri and Luybka Doukovska and Radoslav Yoshinov",

booktitle = "2024 IEEE 12th International Conference on Intelligent Systems (IS)",

address = "United States",

}

Rosner, A , Gegov, A , Hopgood, AA , Lamptey, O , Ouelhadj, D & Da Deppo, S 2024, Machine learning-based classification of extremism using explainable artificial intelligence. in V Sgurev, V Jotsov, V Piuri, L Doukovska & R Yoshinov (eds), 2024 IEEE 12th International Conference on Intelligent Systems (IS). International Conference on Intelligent Systems, Institute of Electrical and Electronics Engineers Inc., 12th IEEE International Conference on Intelligent Systems, IS 2024, Varna, Bulgaria, 29/08/24. https://doi.org/10.1109/IS61756.2024.10705241

Machine learning-based classification of extremism using explainable artificial intelligence. / Rosner, Anna ; Gegov, Alexander ; Hopgood, Adrian Alan et al.
2024 IEEE 12th International Conference on Intelligent Systems (IS). ed. / Vassil Sgurev; Vladimir Jotsov; Vincenzo Piuri; Luybka Doukovska; Radoslav Yoshinov. Institute of Electrical and Electronics Engineers Inc., 2024. (International Conference on Intelligent Systems).

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

TY - GEN

T1 - Machine learning-based classification of extremism using explainable artificial intelligence

AU - Rosner, Anna

AU - Gegov, Alexander

AU - Hopgood, Adrian Alan

AU - Lamptey, Odartey

AU - Ouelhadj, Djamila

AU - Da Deppo, Serge

PY - 2024/10/9

Y1 - 2024/10/9

N2 - This paper presents the first published application of multiple existing machine learning methods to a subset of features taken from the Profiles of Individual Radicalization in the United States (PIRUS) database to predict the feature ‘violent’. The best- performing model in terms of accuracy is the Hist Gradient Boosting model, with an accuracy of 89.06%, which is an improvement of more than 2.5% compared to the benchmark application. Permutation Feature Importance (PFI) and the explanation framework SHAP were then applied to explain the model predictions. Using both of these techniques together allows for a holistic view of both the model’s inner workings and the impact of the features on the results.

AB - This paper presents the first published application of multiple existing machine learning methods to a subset of features taken from the Profiles of Individual Radicalization in the United States (PIRUS) database to predict the feature ‘violent’. The best- performing model in terms of accuracy is the Hist Gradient Boosting model, with an accuracy of 89.06%, which is an improvement of more than 2.5% compared to the benchmark application. Permutation Feature Importance (PFI) and the explanation framework SHAP were then applied to explain the model predictions. Using both of these techniques together allows for a holistic view of both the model’s inner workings and the impact of the features on the results.

KW - Machine Learning

KW - Extremism

KW - PIRUS database

KW - eXplainable AI

KW - SHAP

KW - Permutation Feature Importance

UR - https://www.scopus.com/pages/publications/85208412693

U2 - 10.1109/IS61756.2024.10705241

DO - 10.1109/IS61756.2024.10705241

M3 - Conference contribution

AN - SCOPUS:85208412693

SN - 9798350350999

T3 - International Conference on Intelligent Systems

BT - 2024 IEEE 12th International Conference on Intelligent Systems (IS)

A2 - Sgurev, Vassil

A2 - Jotsov, Vladimir

A2 - Piuri, Vincenzo

A2 - Doukovska, Luybka

A2 - Yoshinov, Radoslav

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 12th IEEE International Conference on Intelligent Systems, IS 2024

Y2 - 29 August 2024 through 31 August 2024

ER -

Rosner A , Gegov A , Hopgood AA , Lamptey O , Ouelhadj D, Da Deppo S. Machine learning-based classification of extremism using explainable artificial intelligence. In Sgurev V, Jotsov V, Piuri V, Doukovska L, Yoshinov R, editors, 2024 IEEE 12th International Conference on Intelligent Systems (IS). Institute of Electrical and Electronics Engineers Inc. 2024. (International Conference on Intelligent Systems). doi: 10.1109/IS61756.2024.10705241

Machine learning-based classification of extremism using explainable artificial intelligence

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Fingerprint

Cite this