The improvement and application of a K-means clustering algorithm
Tao Li, Liu Yin Hong, Hao Jing Yan · 2016
This paper proposes a K-means algorithm with the dynamic adjustable number of clusters. The algorithm uses the improved Euclidean distance formula to calculate the distance between the cluster center and data, by judging whether the distance is greater than the threshold to automatically adjust the number of clusters. Finally, the improved algorithm is applied to intrusion detection system to detect unknown attacks. The test results shows, Compared with traditional K-means algorithm, the K-means algorithm with the dynamic adjustable number of clusters has a better effect.