An Intrusion Detection System Using Machine Learning Algorithm

Security of data in a network based computer system has become a major challenge in the world today. With the high increase of network traffic, hackers and malicious users are devising new ways of network intrusion. In other to address this problem, an intrusion detection system (IDS) is developed which will detect attacks in a computer network. In this research, the KDDCup99 Test datasets is analyzed using certain machine learning algorithms (Bayes Net, J48, Random Forest, and Random Tree) to determine the accuracy of these algorithms by classifying these attacks into their various classes. A constructive research methodology is adopted throughout this research. The experimental results show that the Random Forest and Random Tree algorithms appear to be the most efficient in performing the classification technique on the Test dataset. The experimental tool used is WEKA which is used to perform a correlation based feature selection on the dataset with a Best First search method, and the parameters used for the computation are Precision, Recall and F-measure. Keywords: Intrusion detection system; KDDCup99; Machine learning; WEKA