The Research And Design Of Metasearch Engine On Internet

The Internet has already become the main place to gain and exchange information at present and in the future. However, it is an extremely challenging work to find the satisfied information in a great deal of information on the Internet. The development of the existing traditional search engine, on one hand strengthened its practicability constantly and solved the information search issue partly; on the other hand, it restricted people to obtain the more effective resources, because of its own limitations. It is necessary to design a series of highly effective solutions for obtaining valuable information from the Internet. The metasearch engine is the exactly technique that can meet the needs, which adopts the query ability of these existing search engines, and regards them as a unity, and provids users with an uniform interface and the returning.In this paper, the search engine principles and its current development status were firstly discussed, and the disadvantages of existing traditional search engines were also analyzed. Based on the preceding analysis, we have proposed the future research directions and analyzed the necessity of introducing metasearch engine. Secondly, the metasearch engine principles and its current development status were summarized, and the difficulties of research about the current metasearch engines were pointed out. Moreover we have given some methods for improvement. Then some key algorithms of metasearch engine were discussed, namely component search engines selection algorithms, document selection algorithms and result mergingalgorithms. Based on the above, a metasearch engine model based on Agent was presented, which has adopted the metasearch engine structure, and has regarded Agent as the basic component. The model utilized the autonomy and the coordination of Agents to finish user's personal search. The model framework, key techniques and strategies were also introduced.In this paper, the component search engine selection algorithm and the result merging algorithm were designed, which can help metasearch engine to choose potential and useful component search engines to execute information search. The analyzing of the experiment data demonstrated that some characteristic value impact the selection of component search engine in each tactics. Finally, we have made the evaluation and got conclusion on this model, and have presented the future work.