Research On Tags Recommendation Techniques For Query Intention

With the development of the computer technology, more and more people obtain theinformation which they need from the network. This is not just because of the network ofconvenient or fast, but the network information has been involved in politics, culture,education and many other fields, and can provide adequate information for the needs of thosethey need. The information network has become an important guarantee of socialdevelopment. How users find the relevant information from the mass network information hasbecome an important part of the existing research.For traditional information retrieval techniques, the users search query keywords to meettheir query intention, however, the search returns results contain a large number of irrelevantinformation, from this we can know keywords searching can not provide enough informationto retrieve the relevant documents. Tags based on user-generated, brought us a new way todiscovery users’ query intention. The social tags in Web2.0result from users’ annotationbehavior for pages that they are interested in. The users use tags frequently to describe thesetopics, therefore the tags not only represent the user’s query intention, but also reveal the webinformation best.Based on the research of keywords and tags, the paper presents a research foruser-oriented query intention, it aims at adding the tags which can reflect the user’s real queryintention to the query. Tags as a supplement to the query keywords, not only can make up thedefects of short queries, but also accord to the correlation between the query tags and webcontent sort the related results. If the user just used the keywords, we need to use the querylog to recommend high-quality tag used as a supplement to the query keywords, and returnmore relevant results. Finally, in the experiment, the paper uses several kinds of methods totest and verify the results. The experiments show that the user-oriented query intention searchmethod can better determine the user search interest than the traditional search.