Internal Intrusion Detection System Using Machine Learning

With billions of users connected to the internet today, ensuring cybersecurity has become increasingly critical. Intrusion Detection and Protection Systems (IDPS) represent a new wave of security solutions designed to monitor and respond to harmful activities within a network. This research focuses on a call-level Intrusion Detection and Protection System that employs neighborhood-based procedural grids to observe and analyze user behavior over time, particularly identifying patterns associated with malicious intent. The system builds user profiles to detect deviations from typical usage, enhancing the ability to detect internal threats. The proposed approach is evaluated using forensic techniques and established intrusion detection methodologies. This paper also provides a comprehensive review of existing literature on both general Intrusion Detection Systems (IDS) and Internal Intrusion Detection Systems (IIDS), which utilize real-time data processing and algorithmic analysis to detect threats. Emphasizing the role of cyber analytics, the research presents the development of an IIDS that leverages predefined algorithms to effectively differentiate between legitimate user activity and potential security breaches.