Authors : Mohd Zakree Ahmad Nazri, Nor Emizan Abdul Majid, Azuraliza Abu Bakar and Hafiz Mohd Sarim
Abstract: The increasing rates of intrusion in the network systems has placed many organizations and various institutions in a very vulnerable position of having to face risks which could involve great financial or other losses. A large number of computers have been hacked over the years because some business organizations and other concerns had not taken the essential precautionary measures to protect their networks from cyber-attacks. Since, the frequency of the attacks on the network systems have dramatically increased in number and in the level of destruction over the past years, the Intrusion Detection System has become an important component to ensure the safety for such network systems. This study uses the associative classification algorithm to predict oncoming network intrusions. In order to protect network systems an algorithm is used to detect a variety of relationships and association rules of interest in the flow of traffic incidents. For this research, an attempted network thread data was obtained from a finance company in Malaysia, comprising more than 38,000 records and 20 attributes. The association rule mining is used to discover the sequence of the incidents that enables the model to predict the forth coming incidents in the network systems which in this case are considered as intrusions.
Mohd Zakree Ahmad Nazri, Nor Emizan Abdul Majid, Azuraliza Abu Bakar and Hafiz Mohd Sarim, 2012. Network Intrusion Prediction Using Associative Classification Method. Asian Journal of Information Technology, 11: 153-159.