Can we use fuzzy c-means for feature selection in intrusion detection
Answers
In this work, a new method for classification is proposed consisting of a combination of feature selection, normalization, fuzzy C means clustering algorithm and C4.5 decision tree algorithm. The aim of this method is to improve the performance of the classifier by using selected features. The fuzzy C means clustering method is used to partition the training instances into clusters. On each cluster, we build a decision tree using C4.5 algorithm. Experiments on the KDD CUP 99 data set shows that our proposed method in detecting intrusion achieves better performance while reducing the relevant features by more than 80%.
Answer: We propose Fuzzy Kernel C-Means Algorithm as another method to solve IDS problems that we claim provides better results while combined with Kernel Matrix method to reduce the number of selected data features.
Explanation: