2026-10-17T13:11:17Z https://meral.edu.mm/oai

oai:meral.edu.mm:recid/4431 2021-12-13T00:54:59Z 1582963302567:1597824273898 user-ucsy

Feature Selection in Hybrid Intrusion Detection System Myint, Khin Khattar Kham, Nang Saing Moon With the enormous growth of computernetworks, network security is gaining increasingimportance. Therefore, the role of IntrusionDetection Systems (IDSs) is becoming moreimportant. There are many techniques availablefor intrusion detection. In this paper, a hybridintrusion detection method that integrates ananomaly detection model and a misuse detectionmodel using the one-class SVM and C4.5decision tree is proposed. Despite the inherentpotential of hybrid detection, there are manyissues that highly affect the performance of thehybrid systems such as detection rate, falsepositive rate, memory overhead, time overheadand so on. Moreover, most of the existing IDSsuse all of the features available in the dataset todetect the attack while some of the features areredundant. It is time-consuming and maydegrade the performance of IDSs. Therefore,rough set theory is used in the proposed hybridintrusion detection system to select the mostsignificant features. The experimentation isimplemented in ROSETTA and WEKA toolsusing NSL-KDD dataset. 2016-02-25 http://hdl.handle.net/20.500.12678/0000004431 https://meral.edu.mm/records/4431