# Introduction he World Wide has been dramatically increased due to the usage of internet. The web acts as a medium where large amount of information can be obtained at lower cost. Web mining can be defined as the discovery and analysis of useful information from the World Wide Web data. It is one of the data mining techniques to automatically extract the information from web documents. WWW provides a rich set of data for data mining. The web is dynamic and very high dimensionality. A web page contains three forms of data, structured, unstructured and semi structured data. Data sets available in the web can be very large and occupy ten to hundreds of terabytes, need a large farm of servers. The user are collecting different kinds of information from the global web for both read and writing purpose. In the global web, search is an important activity then only considered to an email. Tremendous growth of web, every second millions of information added in the global web. Users are finding and refinding the web information in the global web everyday [9]. People revisit the information that have ever been come across occasionally or intentionally. Refindng web pages is typically better than to initially finding the webpage. Achieving efficient and accurate information retrieval is a challenging task. Refinding is a common task is difficult when previously viewed information is modified, moved or removed. How information refinding is different from information finding? There is a uncertainty in the later process because users do not know get enough information, while information refinding is a more directed process as users have already seen the information before. Information refinding is not the process of finding again [7]. A general way to support information refinding is to maintain access log [10], recording what users have ever seen based on their revisit frequencies. When refinding, users might prefer to have a search the results prioritized by pages that have been seen before. One way to refinding the information using contextual cues [3][2], inspired from the human memory approach. [8]. The people use lot of keywords to search the information. To remember the keyword after a few months ago what we have seen before it is difficult and time consuming task. Because original queries were wrongly remembered most of that time due to their loss of memory. According to cognitive science literature, human memory is predicted on contextual cues to refinding the information. To get the information for users query exactly even a month or year ago hard to remember that keyword. But the time, place and concurrent activity associated with the happening of that access event may leave a deeper impression. Contextual information could helps as powerful clues to remember the key word. Contextual clues helps to users have seen the already viewed information. Nivethitha (2014) suggested a query analysis for efficient context-based information refinding and page ranking system. Refinding what have done before is a common behavior of human in real life. According to the human natural recall characteristics, users allow to refinding web pages which have seen before. Psycological studies show under which information was accessed can helps as a powerful cue for information recall. Here context including time, place and concurrent activity could serves as a useful information recall clues. In this system not only considered finding the refinding queries. But also implement feedback system, so that webpage can be ranked by the multiple user feedback. Deng et.al. ( 2013) have worked extensively and suggested a effective method for refinding the information fro m the web, they could not remember the 2005) had done a detailed analysis and present an extension to traditional bookmarks called landmarks, a user-directed technique that aid users in returning to specific content within a previously visited webpage. The use of traditional bookmarks allows users to return to a previously visited page, it can be hard to re-find facts within that page. Here we investigate the efficiency of land marks for refinding of information on web-pages. Land mark allow users to mark information on a webpage that they may want to return to a later date by highlighting the text and adding a landmark in the same fashion as they would a favorite in IE. Land marks are not meant as a replacement for the bookmarking facility but as an enhancement that help users return directly to previously visited information, giving context to the marked pages. Hailpern et al. (2011) found that during recall tasks, contextual cues are important component of human memo ry. In this paper they present new interaction technique, pivoting, that allows users to search for contextually related activities and find target piece of information. You Pivot demonstrates how principles of human memo ry can be applied to enhance the search of digital information. Contextual cues could be one way to improve in formation recall in our digital lives. You Pivot used the calendar entry's lifespan as the pivot time period. Time Marks allowing a user to access all activity that was ongoing at a particular moment. Parsons et.al. (2009) extensively worked and suggested a keyword-based information retrieval technique and suggested that the performance can be improved by re-ranking the results based on the context provided by the surrounding terms. A baseline technique was compared against two LSA techniques, and an analysis of the retrieved documents indicated that the re-ranking provided by the LSA techniques significantly improved the efficiency of the retrieved list. However, the participants' performance was not altered by the different techniques. Instead, the findings suggest that, when dealing with a small number of documents, participants will generally access all documents retrieved in a systematic manner. It is therefore hypothesised that the re-ranking technique would be more useful in a significantly larger document collection, where a thorough assessment of all documents is impractical. This study has also emphasized the importance of assessing the impact of individual differences in any information retrieval system. For example, it was found that LSA did improve performance for participants with lower scores on the comprehension test. # Global Journal of Computer # Conclusion We have studied the comparison of various papers of context based information refinding. The aim of this study was how the results of the information retrieval technique to efficiently refinding the web information could be improved by contextual cues shown in above table. # Global ![Divya ? , M. Janga Reddy ? & M. Riyajoddin ? II. Related work . keyword and their related information after a couple of months. It based on the human recall characteristics, allows user to refinding WebPages according to the previous access context. The system was implemented offering some contextual information on the query results. Memory context is also considered in personal information refinding. Based on Inspiration from human memo ry mechanism, the context -based refinding framework was developed.Won et.al. (2009) experimented in their work and identified that most modern web browsers offer web history functionality few people use it to revisit previously viewed web pages. In this paper they developed Contextual Web history (CWH), which improves the visibility of the history feature and greatly reduced the time and effort required to find and revisit webpage.CWH goal is to improve the usability and utility of the history feature in web browsers. CWH provides a richer set of clues about content of the page, including time of visit, visual appearance and text search and quickly find previously visited web page again. Revisiting is a key part of web browsing. Contextual web History gives to understand the right set of basic features to support the process of re-finding information very fast.TasksPotential Memory CuesRecall color, structure, time visited, logos , content, title, url of the web page Recognition Size of the thumbnails MacKay et.al.(](image-2.png "G") :ReferenceAuthorPaper TitleIssuesMethodResultDrawbacksNumber(Refinding)2A.P. NivethithaEfficientTo build recallRe-finderEfficiently revisit ofAll the user notcontext basedbased queryand pagethe web page usinggiventheinformation re-model to re-findrankingcontextual cues andfeedback. So-finding andthe informationmulti user feedback.cannot rankingpage rankingusing contextualthe webpagecues and feedbackproperly.3Tangjian Liang Zhao, Hao Deng, Wang, QingweiRefinder: context-based A informationvisited by the user. To build query-by context model, Context are theA context based Re-finderOn average 15.53 seconds are needed to refinder completeIn Refinder, not implement user feedback forYear 2014Liu, and Ling Fengrefindingpowerfulcuethe refinding requestvisitedwebsystem(place,time,and 84.42 secondspagesconcurrent activity)with other existingforinformationmethodsrefinding.4S. Won, J. Jin, andContextualTodevelopContextualGreatly reduced theIn CWH, re-J. Hong,Web History:ContextualWebWebtimeandeffortfindingaUsing visualHistory(CWH)Historyrequired to refindingwebpage olderandimprovesthethe web pages.than x days toocontextualvisibility of themany pages forcuestohistoryfeaturethe user toimprove Webhelps people findbrowse.Browserpreviously visitedHistoryweb pages.5B. MacKay, M.An EvaluationToimplementLandmarkUsingLandmarksThe users canKellar, and C.of LandmarksLandmark which isrevisit the webpageonlymakeWattersfor Re-findingan extension ofsignificantly faster.landmarks forInformation ontraditionaltextualthe Webbookmarks.information, notLandmark is aexpandthisuser-directedfunctionality totechnique that aidsinclude imagesusers in returningandotherto specific contentmedia.within previouslyvisited webpage6J. Hailpern, N.YouPivot:To allow users toYou PivotUsing You PivotUsersownJitkoff, A. Warr, R.Improvingsearchforgreatly improve thecontextual cuesKarahalios,K.Recallwithcontextual relatedquality and speed ofis difficult toSesek, and N.Contextualactivities(usingrecalldesignShkrobSearchtime marks) andfind a target pieceofdigitalinformation.7Kathryn Parsons,The Use of aThe aim of thisLatentThis study thereforeLSA are unlikelyAgata McCormac,Context-study was toSemantichighlights theto beMarcus Butavicius,Basedexamine whetherAnalysisimportance ofnecessary inSimon Dennis*Informationthe results(LSA)testing the influencerelatively smalland Lael FergusonRetrievalprovided by aof individualdocumentTechniquekeyword baseddifferences on any IRcollectionstechnique wouldsystem, and thebe improvedimportance ofthrough the use oftesting any IR tool ontwo LSAa populationtechniques.that closely reflectsthe intended usersof the system. © 2014 Global Journals Inc. (US) 2 Year 2014 © 2014 Global Journals Inc. (US) © 2014 Global Journals Inc. (US) 4 Year 2014 * A Comparative Study of Context-Based Information Refinding APNivethith DHanirex KPKaliyamurthie An international journal of advanced computer technology 3 4 April-2014 COMPUSOFT. III, Issue-IV * efficiently context-based information re-finding and page ranking APNivethitha 2014 International conference on electrical, communication and computing * ReFinder: A Context-Based Information Refinding System TangjianDeng LiangZhao HaoWang QingweiLiu LingFeng IEEE transactions on knowledge and data engineering 25 9 september 2013 * Contextual Web History: Using Visual and Contextual Cues to Improve Web Browser History SWon JJin JHong Proc. SIGCHI Conf. Human Factors in Computing Systems (CHI) SIGCHI Conf. Human Factors in Computing Systems (CHI) 2009 * An Evaluation of Landmarks for Re-Finding Information on the Web BMackay MKellar CWatters Proc. Extended Abstracts on Human Factors in Computing Systems (CHI '05 EA) Extended Abstracts on Human Factors in Computing Systems (CHI '05 EA) 2005 * You Pivot: Improving Recall with Contextual Search JHailpern NJitkoff AWarr RKarahalios KSesek NShkrob Proc. SIGCHI Conf. Human Factors in Computing Systems (CHI) SIGCHI Conf. Human Factors in Computing Systems (CHI) 2011 * The Use of a Context-Based Information Retrieval Technique" Command, Control, Communications and Intelligence Division Defense Science and Technology Organization KathrynParsons AgataMcCormac MarcusButavicius SimonDennis * LaelFerguson *Ohio State University * Large Scale Analysis of Web Revisitation Patterns EAdar JTeevan STDumais Proc. SIGCHI Conf. Human Factors in Computing Systems (CHI) SIGCHI Conf. Human Factors in Computing Systems (CHI) 2008 * Refinding Is Not Finding Again RCapra MPinney MAPerez-Quinones Aug. 2005 technical report * Integrating Memory Context into Personal Information Re-Finding YChen GJones Proc. Second Symp. Future Directions in Information Access Second Symp. Future Directions in Information Access 2008 * The Re: Search Engine: Simultaneous Support for Finding and Re-Finding JTeevan Proc. 20th 20th * AnnSymp User Interface Software and Technology (UIST) 2007 * Large Scale Query Log Analysis of Re-Finding SKTyler JTeevan Proc. Third ACM Int'l Conf. Web Search and Data Mining(WSDM) Third ACM Int'l Conf. Web Search and Data Mining(WSDM) 2010 * Google Web History