With the expansion of computer network there is a challenge to compete with the intruders who can easily break into the system. So it becomes a necessity to device systems or algorithms that can not only detect intrusion but can also improve the detection rate. In this paper we propose an intrusion detection system that uses rough set theory for feature selection, which is extraction of relevant attributes from the entire set of attributes describing a data packet and used the same theory to classify the packet if it is normal or an attack. After the simplification of the discernibility matrix we were to select or reduce the features. We have used Rosetta tool to obtain the reducts and classification rules. NSL KDD dataset is used as training set and is provided to Rosetta to obtain the classification rules.
GUPTA, NIKITA; SINGH, NARENDER; SHARMA, VIJAY; SHARMA, TARUN; and BHANDARI, AMAN SINGH
"FEATURE SELECTION AND CLASSIFICATION OF INTRUSION DETECTION SYSTEM USING ROUGH SET,"
International Journal of Communication Networks and Security: Vol. 2
, Article 11.
Available at: https://www.interscience.in/ijcns/vol2/iss2/11