An Efficient Approach for Keyword Selection ; Improving Accessibility of Web Contents by General Search Engines

Abstract

General search engines often provide low precise results even for detailed queries. So there is a vital need to elicit useful information like keywords for search engines to provide acceptable results for user's search queries. Although many methods have been proposed to show how to extract keywords automatically, all attempt to get a better recall, precision and other criteria which describe how the method has done its job as an author. This paper presents a new automatic keyword extraction method which improves accessibility of web content by search engines. The proposed method defines some coefficients determining features efficiency and tries to optimize them by using a genetic algorithm. Furthermore, it evaluates candidate keywords by a function that utilizes the result of search engines. When comparing to the other methods, experiments demonstrate that by using the proposed method, a higher score is achieved from search engines without losing noticeable recall or precision.