A Review of Rough Set Theory–Based Naïve Bayes Tree Approaches for Intrusion Detection Systems

The rapid expansion of computer networks, cloud services, and internet-based applications has significantly increased the frequency and complexity of cyber-attacks. Intrusion Detection Systems (IDS) are essential security mechanisms designed to detect unauthorized access, misuse, and malicious activities in networked environments. However, traditional IDS approaches often suffer from limitations such as high false alarm rates, poor scalability, and ineffective detection of novel attacks. To overcome these challenges, intelligent hybrid models integrating feature selection and machine learning classifiers have been widely explored. This paper presents a comprehensive review of Rough Set Theory (RST)–based Naïve Bayes Tree (NB-Tree) approaches for Intrusion Detection Systems. Rough Set Theory is an effective mathematical tool for handling uncertainty and redundancy in high-dimensional datasets, while NB-Tree classifiers combine probabilistic learning with decision tree structures to enhance classification accuracy. The integration of RST with NB-Tree improves detection accuracy, reduces false positives, and lowers computational complexity. This review critically analyzes existing literature, identifies research gaps, outlines a methodological framework, discusses expected outcomes, and highlights future research directions. The study concludes that RST-based NB-Tree models offer an efficient, interpretable, and scalable solution for modern intrusion detection environments.