Intrusion Detection System Based on Frequent Pattern Mining

 Due  to  the  dramatically  increment  of  internet
usage,  users  are  facing  various  attacks  day  by  day.
Consequently, the research area for intrusion detection must
be  fresh  with  new  challenges.  Intrusion  detection  system
includes identifying a set of malicious actions that compromise
the  integrity,  confidentiality,  and  availability  of  information
resources.  The  major  contribution  is  to  apply  data  mining
approach for network intrusion detection system. Among the
several features of data  mining, association rules  mining,FP-
growth algorithm, is used to find out the frequent itemsets of
incoming packets database. Based on these itemsets, anomaly
detection  is  added.  The  system  will  predict  whether  the
incoming data packet is normal or attack. The performance of
proposed system is tested by using KDD-99 datasets.