Effective Pre-Processing Activities in Text Mining using Improved Porter’s Stemming Algorithm

Abstract: Text Databases are rapidly growing due to the increasing amount of information available in various electronic forms. User need to access relevant information across multiple documents. Initial process in Text Mining system is Pre- Processing steps. Our approach to make an effective Pre-Processing steps to save both space and time requirements by using improved Stemming Algorithm. Stemming algorithms are used to transform the words in texts into their grammatical root form. Several algorithms exist with different techniques. The most widely used stemming algorithm is “M.F Porter stemming algorithm. However, it still has certain drawbacks of handling Named Entities. Our paper is to improve its structure by refining with certain constraints, so that improve the Information Retrieval System’s Efficiency. Thus our paper is demonstrate how we can effectively overcome the problem of Named Entity during stemming process.

Keywords: Extraction, NamedEntity, Stemming, StopWordRemoval.