Improved SOM Clustering for Software Component Catalogue

The faceted catalogue is one of most popular methods among software component catalogues. However, the traditional faceted catalogue needs experts to establish and maintain glossary. To solve the problem, SOM clustering is applied in software catalogue to automatically form the glossary. By the improved SOM clustering, faceted descript values are clustered and the centers are taken as faceted terms. In SOM algorithms, distances of components are reduced or extended according to component connectors. So the components in this catalogue are not isolated but have relations with each other. And the problem that clustering result is disturbed by the input order, is also solved in the improved SOM clustering.

[1]  Dieter Merkl,et al.  SELF-ORGANIZING MAPS AND SOFTWARE REUSE , 1998 .

[2]  James C. Bezdek,et al.  Generalized clustering networks and Kohonen's self-organizing scheme , 1993, IEEE Trans. Neural Networks.

[3]  Jiawei Han,et al.  Data Mining: Concepts and Techniques , 2000 .

[4]  Witold Pedrycz,et al.  Computational intelligence in software engineering , 1997, CCECE '97. Canadian Conference on Electrical and Computer Engineering. Engineering Innovation: Voyage of Discovery. Conference Proceedings.

[5]  D. Merkl,et al.  Content-based software classification by self-organization , 1995, Proceedings of ICNN'95 - International Conference on Neural Networks.

[6]  Panos Constantopoulos,et al.  Component classification in the software information base , 1995 .

[7]  Teuvo Kohonen,et al.  Self-Organizing Maps , 2010 .