Research on knowledge search engine based on personalized archives

In the full-text retrieval technology, the traditional keyword information full-text retrieval technology mainly carries out information retrieval for users through the retrieval method matching with the keyword information. It often can only carry out a mechanical information matching for users, which eventually leads to a variety and clutter of retrieval results, Users still need to continue searching in the search results to find their main target search information. Moreover, the retrieval statements entered by users often contain many logical relationships. If these logical relationships are used and added to the additional conditions of retrieval, the accuracy of file retrieval will enter a new level. However, the current file retrieval system often does not support semantic level retrieval, so it is unable to understand the user's retrieval purpose and real intention. This paper improves the accuracy of user search through user preference model and synonym conversion model. In order to increase the accuracy of the search engine.