A Comparative Analysis of Text Classification Algorithms for Ambiguity Detection in Requirement Engineering Document Using WEKA

Abstract

The volume of digital documents is increasing day by day and thus the task of automatic categorization of document is very important for information and knowledge discovery. Classification is the most common method for finding the mine rule from the large databases. Ambiguity is the major problem in Requirement Engineering (RE) documents. Our proposed work uses WEKA text classification technique to identify and classify ambiguity in the RE document. The present study uses different algorithms on the ambiguity detection dataset and on the basis of different statistical measures like accuracy, time, and error rate we find suitable algorithms for this purpose. The main aim of this paper is to do a comparative study of various classification techniques and methodologies and a detailed analysis of different statistical parameters that are used in classification algorithms in order to analyze the quality of classification.