TY - GEN

T1 - A noise-resistant fuzzy c means algorithm for clustering

AU - Chintalapudi, Krishna K.

AU - Kam, Moshe

PY - 1998/1/1

Y1 - 1998/1/1

N2 - Probabilistic clustering techniques use the concept of memberships to describe the degree by which a vector belongs to a cluster. The use of memberships provides probabilistic methods with more realistic clustering than "hard" techniques. However, fuzzy schemes (like the fuzzy c means algorithm, FCM) are often sensitive to outliers. We review four existing algorithms, devised to reduce this sensitivity. These are: the noise cluster (NC) algorithm of Dave (1991), the possibilistic c means (PCM) scheme of Krishnapuram and Keller (1996), the least biased fuzzy clustering (LBFC) method of Beni and Liu (1994), and the fuzzy possibilistic c means algorithm of Pal et al. (1997). We then propose the new credibilistic fuzzy c means (CFCM) algorithm to improve on these methods. It uses a new variable, credibility of a vector, which measures the typicality of the vector to the whole data set. By taking credibility into account CFCM generates centroids which are less sensitive to outliers than other techniques, and closer to the centroids generated when the outliers are artificially removed.

AB - Probabilistic clustering techniques use the concept of memberships to describe the degree by which a vector belongs to a cluster. The use of memberships provides probabilistic methods with more realistic clustering than "hard" techniques. However, fuzzy schemes (like the fuzzy c means algorithm, FCM) are often sensitive to outliers. We review four existing algorithms, devised to reduce this sensitivity. These are: the noise cluster (NC) algorithm of Dave (1991), the possibilistic c means (PCM) scheme of Krishnapuram and Keller (1996), the least biased fuzzy clustering (LBFC) method of Beni and Liu (1994), and the fuzzy possibilistic c means algorithm of Pal et al. (1997). We then propose the new credibilistic fuzzy c means (CFCM) algorithm to improve on these methods. It uses a new variable, credibility of a vector, which measures the typicality of the vector to the whole data set. By taking credibility into account CFCM generates centroids which are less sensitive to outliers than other techniques, and closer to the centroids generated when the outliers are artificially removed.

UR - http://www.scopus.com/inward/record.url?scp=0031633290&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031633290&partnerID=8YFLogxK

U2 - 10.1109/FUZZY.1998.686334

DO - 10.1109/FUZZY.1998.686334

M3 - Conference contribution

AN - SCOPUS:0031633290

SN - 078034863X

SN - 9780780348639

T3 - 1998 IEEE International Conference on Fuzzy Systems Proceedings - IEEE World Congress on Computational Intelligence

SP - 1458

EP - 1463

BT - 1998 IEEE International Conference on Fuzzy Systems Proceedings - IEEE World Congress on Computational Intelligence

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 1998 IEEE International Conference on Fuzzy Systems, FUZZY 1998

Y2 - 4 May 1998 through 9 May 1998

ER -