0.25
0.5
0.75
1.25
1.5
1.75
2
Probabilistic Inference for Graph Classification
Published on Feb 25, 20076548 Views
Graph data is getting increasingly popular in, e.g., bioinfor- matics and text processing. A main dificulty of graph data processing lies in the intrinsic high dimensionality of graphs, namely, when
Related categories
Chapter list
Graph Mining Applications in Machine Learning Problems00:01
Motivations for graph analysis02:28
Graphs03:28
Graph Structures in Biology04:07
Overview05:00
Path Representations & Marginalized Graph Kernels06:32
Marginalized Graph Kernels06:35
Label Path07:57
Path-Probability Vector09:05
Kernel Definition09:48
Computation11:37
Graph Kernel Applications13:35
Strong Points of MGK14:33
Drawbacks of Graph Kernels15:20
Substructure Representation & Graph Mining18:06
Substructure Representation18:16
Graph Mining pt 120:31
Graph Mining pt 221:59
Enumeration on Tree-Shaped Search Space23:15
Tree Pruning24:52
Gspan26:30
Depth First Search (DFS) Code27:02
Discriminative Patterns27:47
Multiclass Version29:21
Summary of Graph Mining30:13
EM-Based Clustering of Graphs (Tsudaand Kudo, ICML 2006)30:58
EM-Based Graph Clustering31:12
Probabilistic Model31:57
Ordinary EM Algorithm32:45
Regularization33:58
E-Step35:36
M-Step36:27
Solution pt 137:28
Solution pt 238:18
Important Observation38:22
Mining for Active Patterns38:57
Experiments - RNA Graphs39:45
Clustering RNA Graphs41:12
Examples of RNA Graphs42:14
ROC Scores42:43
No of Patterns & Time44:24
Found Patterns45:37
Conclusion46:23
Ongoing Work46:58
Multiclass Version (a)50:20