High performance computing RBFN for classification of sizable microarrays

The data retrieved from microarray cover the varieties in its nature, and changes observed with time. The size of gene expression data is huge, which brings computational and analytical challenges while performing classification. The analysis has become an arduous task, especially with the increase in size of commercially available probe sets due to rise in number of features for an experimental set. In this paper, Radial Basis Function Network based on Spark framework (sf-RBFN) is proposed that can run on scalable cluster with multiple nodes. The proposed sf-RBFN classifier can be applied to classify large datasets having size in terms of gigabytes or terabytes. The performance of sf-RBFN is tested with the help of microarray data of various dimensions. A comparative study has been presented on performance and efficiency of both Spark based RBFN classifier (sf-RBFN) and traditional RBFN respectively. Experimental results show that the processing efficiency of the proposed sf-RBFN classifier is significantly superior, as compared to RBFN.