A tweet-centric approach for topic-specific author ranking in micro-blog

  • Authors:

  • Shoubin Kong;Ling Feng

  • Affiliations:

  • Department of Computer Science and Technology, Tsinghua University, Beijing, China;Department of Computer Science and Technology, Tsinghua University, Beijing, China

  • Venue:
  • ADMA'11 Proceedings of the 7th international conference on Advanced Data Mining and Applications - Volume Part I
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

Most users play two roles in micro-blog, namely, author and reader of tweets. Facing diverse users and mass user-generated contents in micro-blog, identifying and ranking influential authors who post topic-specific high-quality contents is a challenge. In this paper, we present a way to measure the quality of tweets, which accordingly determines the influence of their authors. The quality of the tweet is evaluated according to the topic focus degree, the retweeting behavior, and the topic-specific influence of the users who retweet it. In this way, the relationships between two micro-blog users extend beyond the traditional following (i.e., friend -follower ) relationship to have more that are established indirectly and dynamically through tweets. We explore the use of these enriched relationships and present a tweet-centric topic-specific author ranking in micro-blog. To enable timely mass data processing on a daily or even hourly basis, we implement our ranking method using MapReduce framework. Some evaluation experiments have been conducted based on a large-scaled real dataset from Tencent micro-blog, which has the largest number of users (over 200 millions) in China. The result shows that our author ranking approach outperforms the PageRank-based and HITS-based approaches significantly in terms of ranking accuracy and quality.