Feature based Text Classification using Application Term Set

Abstract

In the present world of information, text classification is a more challenging process due to the larger number of training cases and feature set present in text data. One of the most difficult tasks in the text classification problem is high dimensionality of the feature space. As many real world text classifications are not modeled or too difficult to model, this paper aims at the real world text classification approach or model based on one of the properties of David Merrill's First principles of Instruction (FPI). The Objective is to introduce a method to improve text classifications effectiveness, efficiency and accuracy. In this methodology we categorizes the text using a predefined category group by providing them with the proper training set based on the feature of Application phase in FPI. The algorithm involves the Parsing, text categorization and text analysis. General TermsPattern Recognition, Text Mining, et. al.

A Survey on Text Mining Techniques and Methods: A Review Approach

“…In filter method, selection of the subset is done without considering any algorithm which is usually done before processing. In wrapper method, evaluation of feature set is done by considering the algorithm [7]. …”