Query Expansion Based on a Personalized Web Search Model
ABSTRACT A novel query expansion algorithm is proposed in this paper. It is based on a model of personalized web search system. The new system, as a middleware between a user and a Web search engine, is set up on the client machine. It can learn a user's preference implicitly and then generate the user profile automatically. When the user inputs query keywords, more personalized expansion words are generated by the proposed algorithm, and then these words together with the query keywords are submitted to a popular search engine such as Baidu or Google. These expansion words can help a search engine retrieval information for a user according to his/her implicit search intentions. The new Web search model can make a common search engine personalized, that is, throughout personalized query expansion the search engine can return different search results to different users who input the same keywords. The experimental results show the effect and applicability of the presented work for personalized information service of a search engine.
- SourceAvailable from: thuir.cn[show abstract] [hide abstract]
ABSTRACT: The word mismatch problem of various expressions of the same concept between known document resources and user query is one of the main factors that hurt the retrieval performance. This paper proposes a document refinement (DR) approach by expansion and replacement based on semantic relations between words. Different from traditional query expansion technology, this DR approach clusters the information with the same concept, which is closer to human thinking habit of information seeking. Furthermore, an effective real-time DR strategy has been given, which makes the DR approach feasible to applications. Experiments on standard testing set showed that the DR approach made consistent 14% to 23.4% improvements after refinement, and got 16% improvements compared with the corresponding query expansion technologies.
- [show abstract] [hide abstract]
ABSTRACT: This paper describes the work achieved in the first half of a 4-year cooperative research project (ARCADE), financed by AUPELF-UREF. The project is devoted to the evaluation of parallel text alignment techniques. In its first period ARCADE ran a competition between six systems on a sentence-to-sentence alignment task which yielded two main types of results. First, a large reference bilingual corpus comprising of texts of different genres was created, each presenting various degrees of difficulty with respect to the alignment task.Second, significant methodological progress was made both on the evaluation protocols and metrics, and the algorithms used by the different systems. For the second phase, which is now underway, ARCADE has been opened to a larger number of teams who will tackle the problem of word-level alignment.01/1998;
- 01/1984; McGraw-Hill Book Company., ISBN: 0-07-054484-0