Cross-validation for detecting label poisoning attacks: a study on random forest algorithm.

Yasarathna, Tharindu Lakshan; Munasinghe, Lankeshwara; Kalutarage, Harsha; Le-Khac, Nhien-An

Abstract

The widespread adoption of machine learning (ML) algorithms has revolutionized various aspects of modern life. However, their susceptibility to data poisoning attacks remains a significant concern due to their potential to compromise model integrity and performance. This study examines the impact of two types of data poisoning attacks on the Random Forest algorithm. It highlights the vulnerability of ML systems, especially in continual learning settings. We propose a simple yet effective strategy for continual learning ML systems to detect potential label poisoning attacks. This involves observing significant performance changes during model retraining. Experimental evaluation with Random Forest algorithms confirms the efficacy of the strategy in detecting and mitigating label poisoning attacks in continual learning systems.

Citation

YASARATHNA, T.L., MUNASINGHE, L., KALUTARAGE, H. and LE-KHAC, N.-A. 2024. Cross-validation for detecting label poisoning attacks: a study on random forest algorithm. In Pitropakis, N., Katsikas, S., Furnell, S. and Markantonakis, K. (eds.) Proceedings of the 39th International Federation for Information Processing (IFIP) International conference on ICT systems security and privacy protection 2024 (IFIP SEC 2024), 12-14 June 2024, Edinburgh, UK. IFIP Advances in information and communication technology, 710. Cham: Springer [online], pages 451-464. Available from: https://doi.org/10.1007/978-3-031-65175-5_32

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 39th International Federation for Information Processing (IFIP) International conference on ICT systems security and privacy protection 2024 (IFIP SEC 2024)
• Start Date: Jun 12, 2024
• End Date: Jun 14, 2024
• Acceptance Date: Apr 15, 2024
• Online Publication Date: Jul 26, 2024
• Publication Date: Dec 31, 2024
• Deposit Date: Aug 16, 2024
• Publicly Available Date: Jul 27, 2025
• Publisher: Springer
• Peer Reviewed: Peer Reviewed
• Pages: 451-464
• Series Title: IFIP Advances in information and communication technology
• Series Number: 710
• Series ISSN: 1868-4238 ; 1868-422X
• ISBN: 9783031651748; 9783031651779
• DOI: https://doi.org/10.1007/978-3-031-65175-5_32
• Keywords: Data poisoning attacks; Continual learning; Machine learning; Cybersecurity; Random forest
• Public URL: https://rgu-repository.worktribe.com/output/2419015