Survey on Predictive Algorithms to Detect Insider Threat on a Network Using Different Combination of Machine Learning Algorithms
Document Type
Conference Proceeding
Publication Date
2024
Abstract
This study explores the efficacy of predictive algorithms for insider threat detection in organizational networks. Leveraging machine learning and deep learning techniques, the research identifies state-of-the-art models and assesses challenges associated with scalability and imbalanced datasets. Following the Prisma International Standards methodology, 531 articles were initially retrieved, with 59 high-quality articles selected for detailed analysis. The findings reveal a multifaceted approach, with 58.8% proposing new models, 23.5% implementing and evaluating, and 17.6% lacking explicit metrics. SVM and RNNs emerged as frequently used algorithms, reflecting versatility and effectiveness in network traffic analysis for insider threat detection. The study provides insights into current trends, challenges, and potential avenues for future research in the realm of insider threat detection. © 2024 Elsevier B.V., All rights reserved.
Recommended Citation
Femi-Oyewole, Favour; Osamor, Victor Chukwudi; and Okunbor, Daniel, "Survey on Predictive Algorithms to Detect Insider Threat on a Network Using Different Combination of Machine Learning Algorithms" (2024). College of Health, Science, and Technology. 1169.
https://digitalcommons.uncfsu.edu/college_health_science_technology/1169