Monitoring top-k on real-time dynamic social-network graphs

Abstract

This paper presents our solution to 2016 DEBS Grand Challenge. We proposed our original program to efficiently calculate 2 continuous top-k queries on real-time social-network graph data. Our implementation tried to prevent processing of unaffected events by designing the algorithms to efficiently maintain the spare list of candidates of the top-k results. In addition, we improved the efficiency of the state-of-the-art algorithms to speed up the processing of the queries.