Study On The Key Aglorithm Of Verticle Search Engine In Silk Area

The vertical search engine is a branch and extension of the search engine, which can classify information in a certain field from websites. A customized data mining and integration can be implemented and returned to users. The major difference between the vertical search engine and traditional search engine is that it selects the information from websites in a structured way -classify the information while selecting it to better satisfy the search requirements. The vertical search have been applied some special fields, for example chemistry.This paper does some detailed analysis and discussion on the technology of the vertical search engine from the studying and designing angle, and has dealt with the present situation and the trend of search engine at home and abroad. The operating principle has analysed search engine and their every main function. We grasp the two key problems about how to appraise the page subject correlatively and design the high-effect crawlling tactic to suggest a Direction-Question-Search-Machine based on Silk Information.It is the core of a vertical search engine. In the main body, we regarded the designing flow of search engine as the main clue, and focused on discussing the designing and carrying out some typical information identification- method about vertical serach engine, including the algorithms of PageRank and HITS. This paper combines average concept of the HTML page analysis with the Hypertext analysis among the Webs, to satisfy the requirements of search engine system.We adopt the depth-first searching tactics to design a silk information website which has vertical search engine function. And according to the software engineering method, we have studied overall design and realizing process of the silk information search engine. During this process we have discussed an emphasis on the requirement analysis,the design and total structure of web function,database design etc. And we have realized all kinds of algorithms and specific function by code in the design stage as well.A software function test shows that the algorithm of vertical search engine is accurate and steady without the risk of local information resource exhaustion. It supports the search strategy on a fixed site or in a given URL circle. It can also implement some automatic search and download specified information.