K-means Algorithm Based on Improved Krill Herd Algorithm and Calinski-Harabasz Index

Aiming at the problems that the Krill Herd (KH) algorithm is easy to fall into the local optimum, the searchability is weak, and the k-means algorithm is easily affected by the selection of the initial clustering centre, a k-means algorithm based on the improved KH algorithm is proposed. The algorithm is initialized by chaos, dynamic clustering, elite leadership and random mutation strategies to improve the KH algorithm and introduce the optimal clustering number adaptive mechanism, which enhances the comprehensive optimization ability of the algorithm. Six benchmark functions test the improved KH algorithm. The effectiveness of the k-means algorithm based on the improved KH algorithm was tested and verified with UCI machine learning and artificial datasets. The verification results showed that the improved KH algorithm improved based on ensuring a faster convergence speed. Compared with other algorithms, the performance of this algorithm has been significantly improved in all aspects.