Sengupta Nandita, Sil Jaya. Intrusion Detection: A Data Mining Approach

Springer, 2020. — 151 p.Data mining is an integrated process of data cleaning, data integration, data selection, data transformation, data extraction, pattern evaluation, and knowledge presentation. The exponential growth of data opens up new challenges to extracting knowledge from large repositories consisting of vague, incomplete, and hidden information. Data mining research attracted many people working in different disciplines for quite a long period of time. However, the methods lack a comprehensive and systematic approach to tackle several problems in data mining techniques, which are interrelated.
The phrase intrusion detection refers to the detection of traffic anomaly in computer networks/systems with an aim to secure data resources from possible attacks. Several approaches to intrusion detection mechanisms are available in the literature. Most of these techniques utilize principles of machine learning/pattern recognition. Unfortunately, the existing techniques fail to incrementally learn network behavior. The book fills this void. It examines the scope of reinforcement learning and rough sets in handling the intrusion detection problem.
The book is primarily meant for graduate students of electrical, electronics, computer science and technology. It is equally useful to doctoral students pursuing their research on intrusion detection and practitioners interested in network security and administration.