AN IMPROVED FUZZY K-MEANS CLUSTERING ALGORITHM BASED ON WEIGHT ENTROPY MEASUREMENT AND CALINSKI-HARABASZ INDEX
Clustering plays an important role in data mining and is applied widely in fields of pattern recognition, computer vision, and fuzzy control. In this paper, we proposed an improved clustering algorithm combined of both fuzzy k-means using weight Entropy and Calinski-Harabasz index. The advantage of...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Dalat University
2018-07-01
|
| Series: | Tạp chí Khoa học Đại học Đà Lạt |
| Subjects: | |
| Online Access: | http://tckh.dlu.edu.vn/index.php/tckhdhdl/article/view/408 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Clustering plays an important role in data mining and is applied widely in fields of pattern recognition, computer vision, and fuzzy control. In this paper, we proposed an improved clustering algorithm combined of both fuzzy k-means using weight Entropy and Calinski-Harabasz index. The advantage of this method is that it does not only create efficient clustering but also has the ability to measure clusters and rate clusters to find the optimal number of clusters for practical needs. Finally, we presented experimental results on real-life datasets, which showed that the improved algorithm has the accuracy and efficiency of the existing algorithms. |
|---|---|
| ISSN: | 0866-787X 0866-787X |