A K-Means Clustering Algorithm Based on Enhanced Differential Evolution

Abstract

The conventional k-means algorithms are sensitive to the initial cluster centers, and tend to be trapped by local optima. To resolve these problems, a novel k-means clustering algorithm using enhanced differential evolution technique is proposed in this paper. This algorithm improves the global search ability by applying Laplace mutation operator and exponentially increasing crossover probability operator. Numerical experiments show that this algorithm overcomes the disadvantages of the conventional k-means algorithms, and improves search ability with higher accuracy, faster convergence speed and better robustness.