Advanced mountain clustering method
J.W. Lee, Seonghyuk Son, S.H. Kwon · 2002
We introduce the advanced mountain clustering method (AMM), which uses a normalized data space, a Gaussian type mountain function and a destruction method based on a mountain slope. The proposed method is very useful because it needs just one parameter instead of three in the mountain method of Yager and Filev (1994) and finds out cluster centers without any neighboring parasitic cluster centers. In addition, we propose a noniterative selection method for the only parameter /spl omega/. Finally, computer simulation results on numerical examples are presented to show the validity of the proposed clustering method.