Research And Application Of Intuitionistic Fuzzy C-Means Clustering Algorithm

Hesitation degree is always ignored in current intuitionistic fuzzy C-means algorithms. These algo-rithms can’t obtain the hesitation degree in partition matrix, and can’t resolve the intuitionistic clusteringproblem. In this paper, intuitionistic fuzzy partition matrix is re-defined and a new criterion which canobtain the hesitation degree is proposed. For intuitionistic clustering problem, a real intuitionistic fuzzyC-means algorithm (IFCM) is proposed based on multiple objective programming. The soft clusteringalgorithm in intuitionistic environment is further discussed based on the proposed IFCM. Possibilisticintuitionistic partition matrix is re-defined and the possibilistic C-means algorithm in intuitionistic fuzzyenvironment (IPCM) is obtained. Finally, possibilistic intuitionistic fuzzy C-means algorithm (PIFCM)is proposed by combining the advantages of IFCM algorithm and IPCM algorithm. PIFCM algorithmcan resolove the disadvantages in IFCM and IPCM. Each algorithm’s convergence is proved. On theother hand, examples are also used to illustrate the rationality and superiority of the new algorithms insome extent.