0.25
0.5
0.75
1.25
1.5
1.75
2
Social Influence Based Clustering of Heterogeneous Information Networks
Published on Sep 27, 20134017 Views
Social networks continue to grow in size and the type of information hosted. We witness a growing interest in clustering a social network of people based on both their social relationships and their
Related categories
Chapter list
SI-Cluster: Social Influence Based Clustering of Heterogeneous Information Networks00:00
Social Influence00:13
Motivation00:36
Various InformationNetworks involved in SI-Cluster01:59
Activity Network and Influence Network02:45
Problem Definition03:44
Social Influence Propagate across Multiple Influence Networks06:00
Step 1: Reorganize a Heterogeneous Graph into Three Subgraphs06:30
Self-Influence Kernel on Social Graph08:08
Step 2: Compute Self-influence Similarity09:29
Co-influence Kernelon Influence Graph10:43
Step 3: Compute Propagating Co-influence Kernel on Influence Graph10:48
Step 4: Partition Activities into Clusters (area-based)12:14
Propagate Heat Distribution12:18
Step 5.1: Compute Influence Score Based on Co-influence Model13:04
Step 5.2: Classify People Using Influence Propagation13:24
Influence Scores of Authors Based on Conference and Keyword Partitions14:09
Step 6: Compute Co-influence Similarity15:19
Influence-based Similarity15:38
Step 7: Compute Unified Influence-based Similarity16:50
Clustering Objective Function17:08
Parameter-based Optimization18:03
Social Influence Based Graph Clustering20:19
Experimental Evaluation21:33
Cluster Quality Evaluation22:28
Clustering Efficiency22:43
Clustering Convergence23:13
Conclusions23:38
Thank you!24:22