An Hybrid Technique for Data Clustering Using Genetic Algorithm with Particle Swarm Optimization

Data clustering is useful in several areas such as machine learning, data mining, wireless sensor networks and pattern recognition. The most famous clustering approach is K-means which successfully has been utilized in numerous clustering problems, but this algorithm has some limitations such as local optimal convergence and initial point understanding. Clustering is the procedure of grouping objects into disjoint class is known as clusters. So, that objects within a class are extremely similar with one another and dissimilar with the objects in other classes. Firefly algorithm is mainly used for clustering problems, but it also has disadvantages. To overcome the problems in firefly this work used a proposed method of Hybrid K-Mean with GA/PSO. The hybrid method merges the standard velocity and modernizes rules of PSOs with the thoughts of selection from GAs. They compare the hybrid algorithm to the standard GA and PSO approaches. Experimental results show that the proposed method used to reduce the limitations and improve accuracy rate. Keywords— Clustering, K-Mean, Firefly algorithm, Genetic Algorithm (GA), Particle Swarm Optimization (PSO).