Parallel fuzzy rough support vector machine for data classification in cloud environment

Data classification has been actively used for most effective means of conveying knowledge and information to users. With emergence of huge datasets existing classification techniques fail to produce desirable results where the challenge lies in analyzing characteristics of massive datasets by retrieving useful geometric and statistical patterns. We propose a supervised parallel fuzzy rough support vector machine (PFRSVM) for in-data classification in cloud environment. The fuzzy rough set model takes care of sensitiveness of noisy samples and handles impreciseness in training samples bringing robustness to results. The algorithm is parallelized with a view to reduce training times. The system is built on support vector machine library using Hadoop implementation of MapReduce. The algorithm is tested on large datasets present at the cloud environment available at University of Technology and Management, India to check its feasibility and convergence. It effectively resolves outliers' effects, imbalance and overlapping class problems, normalizes to unseen data and relaxes dependency between features and labels with better average classification accuracy. The experimental results on both synthetic and real datasets clearly demonstrate the superiority of the proposed technique. PFRSVM is scalable and reliable in nature and is characterized by order independence, computational transaction, failure recovery, atomic transactions, fault tolerant and high availability attributes as exhibited through various experiments.