| Search engine is a kind of software that can actively search information in world-wide-web and can automatic index and provide searching server. With the quick development of IT, all kinds of websites were built one after another. Each website faces the problem of the internal information acquisition, It is an important problem how to design a general and effectively search mode. The research of the search engine has obtained the enormous development, but currently the search engine still has a few problems as fellow: Search engine's capability of natural language understanding is not enough to fit the needs. Search engine's summary of return to the customer need to improve. Search engine's result evaluation method need to improve.In order to solve these problems, this thesis focus on this aims as follows:1 This paper gave a framework of intelligence search engine and realized parts of functions. This paper gave a model of intelligence search engine, it can fit for any small scaled website to information acquisition.2 We use natural language understanding technique at the interface of the system to increase the intelligence of the search engine.3 This system can provide personalize server to adapt the different request of the different customer, and gave new algorithm of sort query results, and can mine customer's information.4 This paper presented a fast summary extract algorithm.This intelligence search engine has the extensive application value. Now most of the small-scaled websites don't have the capability of information query, some only have the simple function to search information from database, if the metadata changed or the topologic changed the search function will be invalidate. The intelligence search engine can be good solution above problems. |
PDF Full Text Request