Analysis of parts of speech tagging on text clustering

Abstract

Clustering is a machine intelligence which aimed at grouping a set of objects into Subsets or clusters. Clustering text documents into various classifications is a vital advance in indexing, recovery, administration and removal of abundant text data on the Web. In research and development to prove that a new clustering algorithm is efficient, one needs to compare the existing algorithm with the new technique, for which the standard datasets are required. In this paper we have pre-processed the datasets to a standardized format, with an expansion of houses appropriate for a wide range of clustering and related experiments. Our objective is to set up a benchmark document datasets and extract the parts of speech such as verbs, nouns, adverbs, adjectives and etc from the documents of a given dataset and analyze the impact of parts of speech in clustering process.