Abstract
Finding high-quality users to follow is essential for acquiring information in microblogging systems. Measuring user’s quality according to its published posts is effective but also needs a large computation considering the volume and the diversity of the posts. In this paper, we explore using only the posts with URLs, i.e., a subset (∼20 %) of the whole posts, for ranking microblog users and propose an iterative graph based ranking algorithm called UBRank to simultaneously rank users and URLs with the assumption that the importance of users and URLs can be mutually boosted. Experiments based on a Chinese microblog corpus demonstrate the effectiveness of the proposed approach.
Author supplied keywords
Cite
CITATION STYLE
Ye, Y., Li, P., Li, R., Zhou, M., Wan, Y., & Wang, B. (2016). Ranking microblog users via URL biased posts. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10042 LNCS, pp. 85–93). Springer Verlag. https://doi.org/10.1007/978-3-319-48743-4_7
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
