An empirical comparison of four initialization methods for the K-Means algorithm
Pattern Recognition Letters
Learning and Intelligent Optimization
Pattern Recognition Letters
A survey: hybrid evolutionary algorithms for cluster analysis
Artificial Intelligence Review
Hi-index | 0.03 |
We present a new approach for Cluster Analysis based on a Greedy Randomized Adaptive Search Procedure (GRASP), with the objective of overcoming the convergence to a local solution. It uses a probabilistic greedy Kaufman initialization for getting initial solutions and K-Means algorithm as a local search algorithm. We compare it with some typical initialization methods: Random, Forgy, Macqueen and Kaufman. The new approach obtains high quality solutions for the benchmark problems.