Abstract—The interpretation of the quality of clusters and the determination of the optimal number of clusters is still a crucial problem in clustering.
We focus in this paper on multi-SOM clustering method which overcomes the problem of extracting the number of clusters from the SOM map through the use of a clustering validity index. We test the multi-SOM algorithm using real and artificial data sets with different evaluation criteria not used previously such as Davies Bouldin index, Dunn index and silhouette index. The multi-SOM algorithm is compared to k-means and Birch methods. Results show that it is more efficient than classical clustering methods.
Index Terms—Clustering, SOM, multi-SOM, DVI, DB index, Dunn index, Silhouette index.
I. Khanchouch is with the LARODEC Laboratory and High Institute of Management, ISG Tunis, University of Tunis, Tunisia (tel.: +216 50 840 865; e-mail: imen.khanchouch@yahoo.fr).
M. Charrad is with University of Gabes, Tunisia (e-mail: malika.charrad@gmail.com).
M. Limam was with University of Tunis. He is now with Dhofar University, Oman (e-mail: Mohamed.limam@isg.rnu.tn).
[PDF]
Cite: I. Khanchouch, M. Charrad, and M. Limam, "A Comparative Study of Multi-SOM Algorithms for Determining the Optimal Number of Clusters," International Journal of Future Computer and Communication vol. 4, no. 3, pp. 198-202, 2015.
