This document discusses using a combined approach of K-medoids clustering and Naive Bayes classification for intrusion detection. It begins by introducing intrusion detection systems and common data mining techniques used, such as classification and clustering. It then proposes a hybrid approach where K-medoids clustering is first used to group similar data instances, followed by Naive Bayes classification to classify the resulting clusters. The approach aims to maintain high accuracy and detection rates while lowering false alarm rates. An experiment on a prepared dataset showed the proposed approach performed better in terms of accuracy and detection rate with a reasonable false alarm rate.