I. Introduction
Text mining is a process of extracting the unknown, understandable, and ultimately available knowledge from large-scale text data in advance. Text mining is a branch of data mining, whose object is entirely composed of text, is text mining. Therefore, text mining is also known as text data mining or text knowledge discovery, and its main purpose is to extract the interesting, important patterns and knowledge from the unstructured text documents. Text mining can be seen as an extension to database-based data mining or knowledge discovery [1]. Text mining has to deal with those most obscured and unstructured text data, so it has a very close contact with other fields, such as information retrieval, information filtering, automatic summary, text clustering, text classification, natural language processing, artificial intelligence, machine learning, pattern recognition, statistics, visualization and so on. One of the areas in the text mining that is gaining popularity nowadays is text classification or text categorization. In general, text classification is the process of assigning text documents to one or more pre-defined categories based on content similarity [2], [3]. The documents in a collection (or corpus) are usually preprocessed so as to represent them by some numerical measures, before applying supervised learning techniques to create models and subsequently using it to assign predefined category labels to unlabeled documents based on the likelihood inferred [4], [5], [6] and a decent accuracy in classifying documents has been reported. Also some work has been extended to classify documents in web environment. However, the same may be arduous task in question classification.