Interval Attributes Description Based FCM Clustering Algorithm for Noisy Data
Xia Shixiong, Yuee Li, Yong Zheng Zhou · 2008
In allusion to the disadvantages that fuzzy c-means algorithm is sensitivity to noise and possibilistic c-means is easy to generate superposition cluster center, interval attributes description based FCM clustering algorithm is proposed in this paper. Firstly, an interval attributes description model of noisy data is presented. Then a clustering algorithm of interval attributes data based on two ends of interval number implemented by FCM. Finally, the simulations of practical data set are made by the algorithm of this paper and PCM. The results validated the feasibility and efficiencies of the algorithm proposed in this paper.