Improved K-means clustering based on genetic algorithm

The K-means algorithm is widely used because of its reliable theory, simple algorithm, fast convergence and it can effectively handle large data sets. However, the traditional K-means algorithm is sensitive to the initial cluster centers; make the average of all objects in the same class as cluster centers, so clustering results is largely affected by the isolated points. To address the problems, search the initial cluster centers of K-means algorithm used of genetic algorithms, improve the K-means algorithm to reduce the impact of isolated points, the data showed that it has good results.