Abstract:

An improved algorithm was proposed which combined rough set algorithm with fuzzy clustering algorithm. The algorithm took full advantage of lower approximation set and upper approximation set in rough set to solve the problem of uncertain border of fuzzy clustering, getting the result of cluster in lower approximation set and upper approximation set so as to achieve better clustering. It can deal with border issues and complex data issues. The proposed algorithm was applied to researching on cyclodextrin clustering, with the  results showing that compared with K-means clustering algorithm and fuzzy C-means clustering algorithm, improved algorithm has a better clustering effect.

Key words: fuzzy clustering, rough set, clustering, K-means clustering