Microblog user ranking based on PageRank and Hadoop

摘要

Microblog, one of the most popular social networking services, plays an important role in communication and information propagation. With the increasing attention on microblog, mining and analyzing data from microblog is becoming an active area of science. In microblog, the user ranking mostly is only based on the number of followers, but this is not a fair way. In this paper, we propose an improved user ranking algorithm named UserRank based on the traditional PageRank and MapReduce programming model, which processes massive data sets with a cluster of machines on Hadoop platform. The UserRank takes the PageRank value, the forwarding rate, the comment rate and whether the user is verified into consideration, and the experiment result shows that the UserRank algorithm can achieve a more fair and useful ranking result.