In Association with Sri Krishna College of Engineering and Technology, Coimbatore Fourth National Conference on Communication Technology Interventions for Rural and Social Development (ncc –tird 2015)-department of Ece

Network security is a large and growing area of concern for every network. Firewalls are used as a security check point in a network environment. In order to further strengthen the network from illegal access, the concept of Intrusion Detection System (IDS) is important. Clustering technique helps to detect the known as well as unknown types of intrusions. In this paper, K-means clustering with four different multi objectives are used to improve the final solution. Clustering is an unsupervised method and here the 10% of known labels is considered for the clustering. Thus the semisupervised clustering method is posed as an optimization problem and yields the best result compared with unsupervised and supervised method. The number of clusters is chosen as 2 and the fitness function is used to calculate the accurate clustered data. The clustered data is said to optimize sequentially. The Genetic SFLA algorithm is used to optimize the clustered data and the performance is measured using F1-value. The Iris data set from UCI machine learning repository and KDD CUP’99 network data set is considered.