An accelerated mapreduce-based K-prototypes for big data

Abstract

Big data are often characterized by a huge volume and a variety of attributes namely, numerical and categorical. To address this issue, this paper proposes an accelerated MapReduce-based k-prototypes method. The proposed method is based on pruning strategy to accelerate the clustering process by reducing the unnecessary distance computations between cluster centers and data points. Experiments performed on huge synthetic and real data sets show that the proposed method is scalable and improves the efficiency of the existing MapReduce-based k-prototypes method.