Abstract: This Now a days there is very busy schedule and all the users wants to complete their task within a less time. Whenever users fire any Top-k query as an input on World Wide Web then they got number of links as an output. The links which are received by users may contain non-useful information or garbage data. Sometimes these links may contain audios, videos, and Twitter and Facebook comments which is not useful for users. To overcome these problems we develop new proposed Top-k extraction system with better performance. By using this system, user wills saves it’s time to get proper result. By using this system, when user enter Top-k list as a query it will give user direct Top-k list as a result in tabular format. For obtain this result Top-k extraction system we use Top-k extraction algorithm. This system includes all web pages. The web pages is in the structured, unstructured and semi-structured format. Also gives results in less time. This paper is summaries with data which is useful to user that are extract from top-k web pages, which are web pages that web pages includes top k instances of a query which is user fire on web. That’s why top-k lists are highly valuable, give richer data and information.

Keywords: Extraction of web information, Top-k lists, Extraction of list, Web mining, Data mining, Context, Web pages.