As human enters the new century, the rapid development of information society and information industry and the extensive application of information technology have made great contributions to the progress of human civilization, thus search engine technology came into being. Its input is the huge Internet. It uses some strategies to collect and discovery information from Internet. And its output is the sorted and processed data. People can use it to retrieve easily the information they need from the massive repository.
This dissertation firstly reviews the development of the major search engines and the mainly relevant techniques. Currently, the research of WEB search engine is much more than WAP search engine, so the research of WAP search engine is very important. The access to Internet is more popular and more convenient on the mobile phone, so the application of wireless search is more extensive. It meets the needs of people's access to information any time anywhere. In addition, owing to the limitations of mobile phones, WAP search requires higher technical demands than WEB search.
Secondly, this dissertation focuses on the WAP search, which is based on Lucene ,an open source full-text indexing engine and its major research of the search engine’s crawler indexer and searcher. It adds the WAP network traverses functions to achieve the WAP crawler, and then it adds the WML page analysis and indexing functions to achieve the WAP indexer and WAP searcher. Besides, in order to improve the search engine’s accuracy of the result ranking, it analyzes the traditional PageRank algorithm. Because of the lack of the average pass page for the parent page, it links to the right of the value of sub-pages and only considers its direct link to the page. This dissertation proposes the link on the page and analysis of the content and subject of a parent page of non-weight average pass PageRank algorithm, which is relevant to the content of pages linked together with the parent page of the transfer decision weights. The improved algorithm meets the relevance and authority requirements well of the search engine for the ranking result, and effectively reduces the phenomenon of topic drift. It makes the rank of the search result fit the user’s needs.
Finally, it sets up the Lucene-based WAP search engine. And the experimental results show that the improved algorithm can achieve accuracy superior to the traditional algorithmic search.
Keywords: WAP; Search engine; Full text indexing; PageRank
提问者： 阿土仔10000 - 一级