Ranking answering in social networks

Users in social networks can follow and be followed to establish all kinds of cliques. Users´ rankings in followed numbers show their popularities, which is an important measurement for social network analysis. In order to give a fast answering for ranking queries, we proposed four methods, ISH, HDD, BBT and SMH. ISH uses inverted sequential hash, HDD is a combination of blocking and hierarchy, BBT utilizes blocking and B- tree technologies, SMH analyzes the sparse of data in large number space, and adapts sequential storage for low values and multi-hashing storage for high values. Our experiments present their different performance in query time and memory consuming.