Fuzzy Logic Based Consent Neighbor Clustering In High Dimensional Data

High dimensional data clustering arises naturally in a lot of domains, and have regularly presented a great deal with for usual data mining techniques. In this paper, presents an optimal perspective on the problem of Consensus Clustering in high-dimensional data. The proposed method called ―Fuzzy based and kernel mappings with Consensus Neighboring clustering (FKCNC)‖, which takes as key measures of correspondence between pairs of data points. The proposed method is to establish a unified framework for FKCNC on both supervised and supervised data sets. Also, we examine some important factors, such as the clustering quality and assortment of basic partitioning, which may affect the performances of FKCNC. Experimental results on various synthetic and real world data sets demonstrate that FKCNC is highly efficient and is equivalent to the state-of-the-art methods in terms of clustering index quality. In addition, FKCNC shows high robustness to incomplete basic partitioning with many anomaly values.