A New Fuzzy Possibilistic Cmeans Clustering Algorthm Based On Dynamic Time Warping Distance

Most works in present field are centered on the descrip- tion and make use of a distance among sequence of as- pects. A measure called Dynamic Time Warping (DTW) appears to be presently the major related for a huge panel of applications. This work is about the use of DTW in data mining algorithms, and focus on the calculation of an average of a set of sequence. Averaging is sary tool for the investigation of data. For an neces- illustration, the K-means clustering algorithm frequently calculates such an average, and requests to offer a explanation of the group it forms. Averaging is now critical steps, be sound in arrange to construct algorithms fectly. This work performs the technique Possibilistic C-Mean (FPCM) algorithm which the clustering accuracy. Experimental results ous that the FPCM advance create better FCM and MFCM clustering algorithms. which must work per- of Fuzzy enhances make obvi-