ML POWERED INSIDER THREAT DETECTION IN ORGANISATION

Abstract

Insider threats pose a significant challenge to organizational security by exploiting legitimate access to sensitive data and systems. Traditional security measures often fail to detect these threats, highlighting the need for advanced solutions. This study explores the use of machine learning models—Convolutional Neural Networks (CNN), Random Forest (RF), and Decision Trees (DT)—to detect insider threats by analyzing historical behavioral data, such as login activity, data access, and communication logs. While CNNs excel at identifying complex patterns, RF and DT models provide strong classification accuracy and transparency. The models' effectiveness is evaluated using metrics like accuracy, precision, recall, and F1-score, ensuring an optimal balance between detection and false positives. Integrating machine learning with cybersecurity practices enhances threat detection and strengthens organizational security.