The paper proposes a user interest model based on user query keywords expanding, using information contained in an ontology for agriculture search engine, so more personalized search results can be returned by search engine with user interest model. Every user interest model for each user contains several sets of concepts, which are expanded from keywords submitted by user in the past period of time based on agricultural category ontology, and pruned based on user browsing behavior to reduce the model size to accelerate the response of search engine. Finally, the approach has been successfully implemented in a module of agriculture search engine SOUNONG1 which helps the system improve the relevance of responses to user queries.
展开▼