Understanding the topical sense of user queries is a problem at the heart of Web search. Successfully mapping incoming user queries to topical categories, particularly those for which the search engine has domain-specific knowledge, can bring improvements in both the efficiency and the effectiveness of Web search. Much of the potential for these improvements exists because many of today’s search engines, both for the Web and for enterprises, often incorporate the use of topic-specific back-end databases when performing general Web search. That is, in addition to traditional document retrieval from general indices, they attempt to automatically route incoming queries to an appropriate subset of specialized back-end databases and return a merged listing of results, often giving preference to the topic-specific results. This strategy is similar to that used by metasearch engines on the Web. The hope is that the system will be able to identify the set of back-end databases that are most appropriate to the query, allowing for more efficient and effective query processing.
The Web information retrieval tools available only make use of the textual information, while they ignore the link information that could be very valuable. The motive of Web structure mining is to create structural summary of the Web site and Web page, Web content mining primarily focuses on inner-document structure, while Web structure mining tries to find the link structure of the hyperlinks at the inter-document level. On the basis of topology of the hyperlinks, Web structure mining will categorize the Web pages to generate the information, based on the similarity and relationship between different Web sites.
|IEEE Base paper|
|Doc||Complete Project word file document|
|Read me||Complete read me text file|
|Source Code||Complete Code files|