期刊名称：International Journal of Electronics, Communication and Soft Computing Science and Engineering
印刷版ISSN：2277-9477
出版年度：2015
卷号：4
期号：Special 2
出版社：IJECSCSE
摘要：Interaction of the end user with information retrievalsystem requires to execute a correct query to the system which notalways feasible and if user is non-technical then formation ofcorrect query will be difficult. So in this paper we are suggestingis that users’ search should be satisfied in minimum time andaddition to this should consume a less time and system should beuser friendly and accurate. Main focus of our proposed system isif any user is entering a wrong or incorrect query then also wewill try to correct it first of all and present the ‘n’ possibleoutcomes from that incorrect query. Query reforming of search isintended to address the problem of inconsistency of terminology.For example, if you have a query “TOI” and the document onlycontains “New York Times”. Then the query and the documentdo not match well and the document will not be RANKED high.Attempts to transform query reformulation “TOI” to “Times ofIndia” and thus make a better matching between the query anddocument. In the task, given a query one needs to generate allsimilar queries from the original query. Therefore here we areachieving the usability nature of the system and our systembecame user friendly as it does not insist end user to give correctquery only. Though user to enter an incorrect query by applyingthe pruning strategy which will guarantee to generate the ‘n’suitable string or queries which will be possibly relevant with theuser’s query. Proposed method is Applied to correct errors inspelling of queries as well as reformulation of queries in Websearch. Experimental results on large scale data show that theProposed approach is very accurate and efficient Improving uponExisting methods in terms of accuracy and efficiency in differentsettings.
关键词：String Transformation; Log Linear Model; Spelling;Check; Query Reformulation
