Ensemble Based Fuzzy with Particle Swarm Optimization Based Weighted Clustering (Efpso-Wc) and Gene Ontology for Microarray Gene Expression

Data clustering proves to be a useful data mining approach for finding the sets of matching objects existing in the dataset. Scalability to manage massive volumes, reliability towards inherent outlier data and validity of clustering outcomes include the important issues in any data clustering technique. With the aim of addressing these problems, an Ensemble based fuzzy with Particle Swarm Optimization based Weighted Clustering (EFPSO-WC) technique that is extensively parallel and distributed in each stage, is introduced in this research work. Here Gene Ontology (GO) can be utilized for establishing the weight owing to the biological relevance exhibited by genes and its optimization is performed employing PSO. In the newly introduced work, Ensemble integrates different clustering outcomes achieved from fuzzy clustering, Fuzzy Weighted Clustering (FWC) and FPSO-WC of a group of objects into one integrated assorted clustering, frequently known as the harmony solution. This clustering can be utilized for the generation of more reliable and balanced clustering outcomes in comparison with a single clustering technique, carry out distributed computing under strict conditions or sharing information. In addition, the effectiveness of the newly introduced EFPSO-WC approach in terms of scalability and reliability was the compared with recently performed researches on the same subject. In all of the stated assessment analysis, the proposed technique performed better than the works carried out recently on the same datasets.