A distributed Arabic text classification approach using latent semantic analysis for big data

摘要

Recently, big data have special concerns from researchers, this due to the valuable information can be collected from it. LSA has an effective performance in classification, and information retrieval, since it deals with the semantics of the words. In this paper, we proposed a distributed text classification approach based on LSA, and Cosine Similarity, and can be applied to big data. The proposed approach reduced the time required for classification, and outperform the related work in terms of precision, and recall.