A rapid egocentric search scheme using authority estimation in blog space

Abstract

Purpose - The purpose of this study is to improve the egocentric search speed for important documents in neighbouring blogs. Design/methodology/approach - This paper presents a rapid egocentric search scheme that narrows down the search space to more important blogs. To determine which blogs are more valuable among a user's neighbouring blogs, a heuristic function is developed that predicts the authority scores on the basis of the local information of the blog. The proposed approach improves the speed of the egocentric search process and the quality of retrieved documents. Findings - A blog is a new medium that is receiving considerable attention. Its links enable one to acquire information about social relations between bloggers in a blog space, and these relations reflect bloggers' interests. Therefore, the ability to search documents in linked blogs is significant for bloggers. An egocentric search method is proposed to search for documents in such neighbouring blogs. However, it takes considerable time to find the most valuable documents in a user's neighbouring blogs when many blogs are linked to that user's blog. Originality/value - This study shows that the number of neighbouring blogs, which are linked to a blog with trackbacks and comments, is important for estimating the authority of a blog. In the experimental results this method performs about five times faster than the egocentric search using a breadth-first search strategy in searching for the top 5 per cent of the most important documents in the neighbouring blogs.