Evaluation of a New Algorithm for Determining User Search Objectives Using Feedback Sessions

Abstract– When someone enters a search query into a search engine that is both general and ambiguous, it is possible that they have one of a number of different search goals in mind at the time they are doing so. By applying inference and conducting an analysis of user search goals, it is possible to significantly enhance both the quality of the user experience and the degree to which search engine results are relevant to the user’s query. We present a novel approach to determining the query goals of users by performing analysis on the query logs of search engines, much like how we do it for this study. Specifically, we focus on the goals that users are trying to accomplish. In order to provide a methodology for determining the various user search intents that are associated with a query, we will first cluster the various feedback sessions that have been recommended. This will allow us to provide a methodology for determining the various user search intents that are associated with a query. The data that users provide about their click-through activity is used in the construction of feedback sessions that accurately reflect the informative needs of the users. Second, we propose an innovative approach to the production of pseudo-documents, which will result in a change to the manner in which feedback sessions are portrayed for the purpose of clustering. This is done in order to ensure that the results are as accurate as possible. As a conclusion, we would like to propose a new statistic that we will refer to as the “Classified Average Precision (CAP)” as a means of determining how well one is able to ascertain the search intentions of a user. We will use this statistic to determine how well one can predict the results of a search. We present the findings of our studies, which consisted of an analysis of the click-through logs that were produced by commercial search query users, in order to demonstrate that the strategies we have suggested are effective. These findings were obtained by conducting the study.