Efficient text pattern mining and clustering approach for record retrieval using gso based prefix span and improved k m eans algorithm

Abstract

Data mining analyses a large number of observational data sets,finds unsuspected relationships and summarizes the data in novel ways thatare both understandable and useful for the user. The wide-spread use ofdistributed information systems leads to the construction of large datacollections in various fields. Many data mining techniques have beenproposed for mining useful patterns in text documents. However, how toeffectively use and update discovered patterns is still an open research issue,especially in the domain of text mining. Since most existing text miningmethods adopted term-based approaches, they all suffer from the problems ofpolysemy and synonymy. Over the years, people have often held thehypothesis that pattern (or phrase)-based approaches should perform betterthan the term-based ones, but many experiments do not support this newlinehypothesis.The proposed method is performed on the records, based on the twomain phases, which are training and testing phases. In the training phase: 1)applying prefix span algorithm, 2) length and width constraints, 3) Optimalmining via Group Search Optimization (GSO). We first present the concept ofprefix span, which detects the frequent pattern using prefix tree. Based on thisprefix tree, length and width constraints are applied to handle restrictions. newline newline