PROJECT TITLE :

QMSampler: Joint Sampling of Multiple Networks with Quality Guarantee - 2018

ABSTRACT:

As a result of On-line Social Networks (OSNs) have become increasingly vital within the last decade, they need motivated a great deal of analysis on Social Network Analysis (SNA). Currently, SNA algorithms are evaluated on real datasets obtained from massive-scale OSNs, that are usually sampled by Breadth-First-Search (BFS), Random Walk (RW), or some variations of the latter. But, none of the released datasets provides any statistical guarantees on the difference between the sampled datasets and the bottom truth. Moreover, all existing sampling algorithms solely specialize in sampling a single OSN, however every OSN is truly a sampling of a complete social network. Hence, even if the full dataset from one OSN is sampled, the results might still be skewed and might not absolutely replicate the properties of the whole social network. To address the higher than issues, we tend to have created the primary try to explore the joint sampling of multiple OSNs and propose an approach referred to as Quality-guaranteed Multi-network Sampler (QMSampler) that may jointly sample multiple OSNs. QMSampler provides a statistical guarantee on the distinction between the sampled real dataset and the bottom truth (the right integration of all OSNs). Our experimental results demonstrate that the proposed approach generates a a lot of smaller bias than any existing technique. QMSampler has conjointly been released as a free download.