Reduced Feature Selection of KDDCUP99 Dataset Using Entropy, Gain and SVM Classifier

Computer Science and Engineering

Information about ensuring safety in the private sector or the government has become a need. Host intrusion detection systems monitor malicious activities and the management station is a technique that generates reports. Intrusion detection system, the availability of an attack and to protect the integrity of the data used for the detection of attacks. IDS detect intrusions using data mining techniques & other software techniques. The intrusion detection technique can efficiently expand the scope of defense of system. In this work we aim to improve efficiency for intrusion detection system. There are two phases in certain ways, in the first phase, we are using decision tree and SVM classifiers for classification of data and the second phase, we boost both the decision tree and SVM classifiers, and detect intrusion more than a single class classifier system. We are using KNN (k- nearest neighbor) classifier for misclassification data sets to improve detection rate. The kddcup99 dataset is used as a simulation set. KDDCUP 1999 benchmark dataset is used for testing the proposed algorithm and the results are promising and more important, especially high sensitivity, specificity and accuracy to create a model to achieve, that outperforms the existing methods are presented. The result shows that our proposed approach achieves better precision and detection rate by using KNN.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.