Paper Details

Abstract: With the expansion of the Internet, the number of assaults has skyrocketed, and Intrusion Detection Systems (IDS) have emerged as a critical component of information security. The goal of an intrusion detection system (IDS) is to assist computer systems in dealing with assaults. This anomaly detection system builds a database of typical behavior and deviations from it, which it uses to trigger when intrusions occur. IDS model is divided into two types based on the data source: host-based IDS and network-based IDS. Individual packets passing over the network are monitored in network based IDS, whereas actions on a particular computer or host are studied in host based IDS. The feature selection aids in the reduction of categorization time. It has been suggested and implemented in this work to successfully detect assaults. A novel feature selection method based on Information Gain Ratio dubbed Optimal Feature Selection has been developed and implemented for this purpose. The KDD Cup dataset is used to select the best number of features using this feature selection technique. In addition, the data set was effectively classified using two classification techniques: Support Vector Machine and Rule Based Classification. This method is highly effective at identifying DoS assaults and lowering the number of false alarms. The suggested feature selection and classification methods let the IDS identify assaults more effectively.

Intrusion Detection, Information Gain, Feature Selection Technique, Classification