0.25
0.5
0.75
1.25
1.5
1.75
2
Fast Direction-Aware Proximity for Graph Mining
Published on Aug 14, 20076722 Views
In this paper we study asymmetric proximity measures on directed graphs, which quantify the relationships between two nodes or two groups of nodes. The measures are useful in several graph mining t
Related categories
Chapter list
Fast Direction-Aware Proximity for Graph Mining00:03
Proximity on Graph00:20
Edge Direction w/ Proximity00:41
Motivating Questions (Fast DAP)01:12
Roadmap pt 101:31
Defining DAP: Escape Probability01:37
Escape Probability: Example01:59
Escape Probability is Good, but…02:11
Issue 1: \"Degree-1 Node\" Effect02:19
Universal Absorbing Boundary02:41
Introducing Universal-Absorbing-Boundary03:00
Issue 2: Weakly Connected Pair03:10
Practical Modifications: Partial Symmetry03:52
Roadmap pt 204:04
Solving Escape Probability: [Doyle+]04:09
Solving DAP: [Doyle+]05:07
Solving vk (j, i) [Doyle+]05:18
Transition Matrix05:26
Solving DAP (Straight-Forward Way) 05:36
Challenges05:45
FastAllDAP07:06
FastAllDAP: Observation07:15
FastAllDAP: Rescue07:51
FastAllDAP: Theorem08:13
FastAllDAP: Algorithm08:26
FastOneDAP 08:42
FastOneDAP: Observation pt 108:51
FastOneDAP: Observation pt 209:35
FastOneDAP: Observation pt 310:05
FastOneDAP: Iterative Alg.10:15
FastOneDAP: Property10:27
Roadmap pt 310:42
Datasets (All Real)10:45
We Want to Check…10:57
Link Prediction: Existence pt 111:06
Link Prediction: Existence pt 311:32
Link Prediction: Direction11:35
Efficiency: FastAllDAP12:34
Efficiency: FastOneDAP12:52
Roadmap pt 412:58
Conclusion (Fast DAP)12:59
More in the Paper…13:23
Cupid Uses Arrows, So Does Graph Mining!13:39
Fast Direction-Aware Proximity for Graph Mining (a)13:53