An Algorithm Of Search Engine Based On Text Clustering

With Internet greatly expanding, the scale of the information has been larger than before. It's important that how to find the most interesting information users needed. The portal websites tried to make their categories more reasonable and more efficient and provided more services. When facing millions of information, however, they cannot satisfy users'requirement. At the moment, the search engine was considered as the one.At first, the paper provided an introduction about the present situation of search engine, the studying in the whole world and the fundamental principle and problem of text retrieve based search engine. And then, it described some classic clustering algorithm and the experiments about these algorithms which could greatly show the performance. According to the data from the experiments, the paper approached an algorithm named text clustering based search engine algorithm. It based on DBSCAN and brought some improvements that made a better performance. This search engine had much higher precision than text retrieve based one and was more utility value.