K-strings algorithm, a new approach based on Kmeans

K-means is a popular clustering algorithm which is widely used in anomaly-based intrusion detection. It tries to classify a given data set into k (a predefined number) categories. However, to apply to a high dimensional dataset, we believe that the calculated distance of the multitude of different attributes along with diverse ranges, that depend on a sole center point are not fair and limit the chance to generate good quality clusters. Aiming to cluster a high dimensional dataset more effective, we propose K-string clustering algorithm in this paper. In which, we obtain a set of center points, that play the role as the backbone of a cluster, instead of a unique centroid to generate clusters. To evaluate we use fitness function, and absolutely the output clusters shown a competition results when comparing to K-means algorithm and Genetic Algorithm [6].