The Study Of Retrieval Method And Searcher Of Internet Search Engine

With the rapid development of Internet technology, the volume of online information into exponential growth, and how to effectively retrieve information has become the current important research topics. In the search engine system, the choice of methods in of information retrieve affects the search results in a large extent and therefore how to provide an effective search and retrieval mechanisms become a hot engine technology research. The article first introduces the relevant technical information retrieval, content analysis based on the search algorithm, based on the analysis of the super-link search algorithms and retrieval algorithms and integration characteristics summed up the traditional retrieval algorithms exist some problems. And against some of these issues and proposed a super-links and links to describe files based on the choice of expanding retrieval algorithms (SAHITS algorithms). According to the phenomena of HITS algorithms only paying attention to the super-link analysis but completely ignoring the text content of a theme to easily yield a few shortcomings such as drift (Topic Drift). SAHITS does three improvements through selection of the collection of roots, the root sets selective expansion and super links relevance. Finally, this article realizes a HITS and SAHITS algorithm comparison system using Java language on the basis of the improvements made by the algorithm. Experiments have shown that SAHITS algorithms have better performance.