Paper
A new K-means Initial Cluster Class of the Center Selection Algorithm Based on the Great Graph
- Authors:
- Zhao Jianyang; Zhou Haiyan
- Abstract
- K-means algorithm, which is simple and fast, with a more intuitive geometric meaning, has been widely used in pattern recognition, image processing and computer vision, and achieved satisfactory results. But K-means algorithm is executed to be pre-determined initial cluster center of class, the initial cluster centers class selected have a direct impact on the final clustering results. K-means Initial cluster class of the center selection algorithm based on the great group is presented. The method compared to the other initial cluster center of class selection algorithm, significantly improves the clustering effect.
- Keywords
- Data Mining; Data Clustering; Initial Cluster Center; Great Group
- StartPage
- 17
- EndPage
- 20
- Doi