The key idea of this paper is that although the clicked documents are not always relevant to users’ queries, the snippets which lead them to the click most probably meet their information needs. Instead of finding the most similar terms previous users queried, we focus on how to detect users’ actual information need based on their search behaviors. State-of-the-art researches prove that the use of users’ behavior information helps to improve query recommendation performance. Query recommendation helps users to describe their information needs more clearly so that search engines can return appropriate answers and meet their needs. We present a model for search engine queries and a variety o f quasi-similarity measures to retrieve relevant queries. The a p- proach is based on the notion o f quasi-similarity b etween qu eries since full similarity with an un satisfactory qu ery would lead to d isappointment. Assuming that every search qu ery can b e e xpressed d ifferently and that other users with similar information n eeds could have already expressed it better, the system makes use of collaborative knowledge from different search engine users to recommend n ew w ays of expressing the same information n eed. This paper presents a method for building a system for automatically suggesting similar queries when results for a query are not satisfactory. Of- ten, users try d ifferent queries until they are satisfied with the results. On the other hand, it is also true that finding the appropriate query for the best search engine result is not a trivial task. Many agree that t he relevancy o f current search engine results needs significant improvement. In particular, our MVMM approach, consistently leads the pack, making it an effective and practical approach towards Web query recommendation. Results show that the sequence-wise approaches significantly outperform the conventional pair-wise ones in terms of prediction accuracy. Extensive experiments were conducted to benchmark our sequence prediction algorithms against two conventional pairwise approaches on large-scale search logs extracted from a commercial search engine. Different query sequence models were examined, including the naive variable length N-gram model, variable memory Markov (VMM) model, and our proposed mixture variable memory Markov (MVMM) model. In this paper, we propose a novel "sequential query prediction" approach that tries to grasp a user's search intent based on his/her past query sequence and its resemblance to historical query sequence models mined from massive search engine logs. Building a good Web query recommendation system, however, is very difficult due to the fundamental challenge of predicting users' search intent, especially given the limited user context information. Web query recommendation has long been considered a key feature of search engines. The performance of the system shows that the synonym based approach giveīetter and effective recommendation for all queries as compared to previous methods. Synonym based method ranks the clicked URLs at the top of the result based Here for given query recommendation the synonyms areĮxtracted on line. Moreover user preferences can be used to build the user profile which will help Therefore in addition of history and snippets with synonyms Various methods based on history of users and snippets to retrieve the information. Search engine can return appropriate result to meet users’ information needs. Query recommendation can be used to help user to state exactly their information Search engine sometime fails to understand user Recently, growth of internet has been increased for information retrieval though it is difficult toĮxtract the relevant information in less time.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |