Machine Learning–Based Terrorism Forecasting Using XGBoost on Large-Scale Global Data

Mahua Banerjee; Amrita Priyam; Umesh Prasad; Soumitro Chakarvarty; Satyam Kumar; Rishi Raj Choudhary

doi:10.2991/978-94-6239-799-6_16

<Previous Article In Volume

Machine Learning–Based Terrorism Forecasting Using XGBoost on Large-Scale Global Data

Authors

Mahua Banerjee¹, Amrita Priyam², Umesh Prasad³, Soumitro Chakarvarty⁴^{, *}, Satyam Kumar⁵, Rishi Raj Choudhary⁵

¹Assistant Professor, Department of Computer Science, BIT Mesra, Lalpur Campus, Ranchi, India

²Associate Professor, Department of Computer Science, BIT Mesra, Lalpur Unit, Ranchi, India

³Assistant Professor, Department of Computer Science, BIT Mesra, Lalpur Campus, Ranchi, India

⁴Assistant Professor, Department of Management Studies, BIT Mesra, Lalpur Unit, Ranchi, India

⁵MCA Student, Department of Computer Science, BIT Mesra, Lalpur Unit, Ranchi, India

^*Corresponding author. Email: soumitro@bitmesra.ac.in

Corresponding Author

Soumitro Chakarvarty

Available Online 30 September 2026.

DOI: 10.2991/978-94-6239-799-6_16 How to use a DOI?
Keywords: Global Terrorism Prediction; XGBoost; Machine Learning; Geospatial Analysis; FastAPI; PostgreSQL; PostGIS; Interactive Dashboard; Severity Estimation
Abstract: This study offers a Global Terrorism Prediction System that uses the Global Terrorism Database (GTD) to identify trends in terrorist activity using XGBoost, a sophisticated ensemble machine learning method. In order to determine terrorist risk levels based on past characteristics including attack type, geography, target category, weapon type, and temporal trends, the system combines data processing, predictive modeling, and interactive visualization. For spatial querying, data is stored in PostgreSQL with PostGIS, and a FastAPI backend offers prediction endpoints that produce GeoJSON for a dashboard that uses Streamlit. The feature significance and partial graphs are used by the model to provide probabilistic insights particularly for academic and research purposes. The model limits and reporting biases are incorporated on ethical issues. XGBoost Model is superior than baseline models like Random Forest and regular Gradient Boosting, which shows 0.49–51 R² in severity prediction. This study also outlines the significance of machine learning and geospatial analytics in global security studies.
Open Access: Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

<Previous Article In Volume

Volume Title: Proceedings of the International Conference on Emerging Trends and Technologies in Applications of Computer Science (ICETTACS 2026)
Series: Advances in Intelligent Systems Research
Publication Date: 30 September 2026
ISBN: 978-94-6239-799-6
ISSN: 1951-6851
DOI: 10.2991/978-94-6239-799-6_16 How to use a DOI?
Open Access: Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Cite this article

ris enw bib

TY  - CONF
AU  - Mahua Banerjee
AU  - Amrita Priyam
AU  - Umesh Prasad
AU  - Soumitro Chakarvarty
AU  - Satyam Kumar
AU  - Rishi Raj Choudhary
PY  - 2026
DA  - 2026/09/30
TI  - Machine Learning–Based Terrorism Forecasting Using XGBoost on Large-Scale Global Data
BT  - Proceedings of the International Conference on Emerging Trends and Technologies in Applications of Computer Science (ICETTACS 2026)
PB  - Atlantis Press
SP  - 209
EP  - 227
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-799-6_16
DO  - 10.2991/978-94-6239-799-6_16
ID  - Banerjee2026
ER  -

download .riscopy to clipboard