Improved Relevance Ranking in WebGather

The amount of information on the webis growing rapidly, and search engines that rely on keyword matchingusually return too many low quality matches. To improve search results,a challenging task for search engines is how to effectively calculate arelevance ranking for each web page. This paper discusses in what ordera search engine should return the URLs it has produced in response to auser's query, so as to show more relevant pages first. Emphasis is givenon the ranking functions adopted by WebGather that take link structureand user popularity factors into account. Experimental results are alsopresented to evaluate the proposed strategy.