Pattern recognition using neural networks that learn from fuzzy rules

Clustering methods have been used extensively in computer vision and pattern recognition. Fuzzy clustering has been shown to be advantageous over crisp (or traditional) clustering in that total commitment of a vector to a given class is not required in each iteration. Recently fuzzy clustering methods have shown spectacular ability to detect not only volume clusters, but also clusters which are actually “thin shells”, i.e. curves and surfaces. Most analytic fuzzy clustering approaches are derived from the fuzzy C means (FCM) algorithm. The FCM uses the probabilistic constraint that the membership of a data point across classes sum to 1. The memberships resulting from FCM and its derivatives, however, do not always correspond to the intuitive concept of degree of belonging or compatibility. Moreover, the algorithms have trouble in noisy environments. In this paper, we cast the clustering problem into framework of possibility theory. In this paper we introduce a comparative study between clustering using unsupervised learning and possibilistic clustering approach