Абстрактный

A New Hybrid Scalable Parallel Clustering Approach For Large Data Using FCM And FA

Juby Mathew, Dr. R Vijayakumar

Clustering is an unsupervised learning task where one seeks to identify a finite set of categories termed clusters to describe the data. we try to exploit computational power from the multicore processors. we need a new design on existing algorithms and software. This research work analyzes about the performance of parallel k means algorithm and based on this algorithm ,we propose a new parallel architecture combined with PKM,FCM and FA algorithm. Here, the parallel architecture will be developed by including the process like, splitting the input data, clustering each subset of data and merging to optimal final clustering.Clustering of subset of data FCM used and optimal merging process we apply FA. Firefly-based clustering is a recent method which proves better for optimal clustering finding. The experimental results show that the performance of modified parallel algorithmis better than the parallel k-means algorithm.In order to utilize the intrinsiccapabilities of a multi-core processor the software application must be able to execute tasks in parallel using all available CPUs. To achieve this we can use fork/join method in java programming