Probabilistic Modeling and Machine Learning in Structural and Systems Biology

Probabilistic Inference for Graph Classification

author: Koji Tsuda, Max Planck Institute for Biological Cybernetics

Description

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 a graph is represented as a binary feature vector of indicators of all possible sub- graphs, the dimensionality gets too large for usual statistical methods.

Categories

Top: Computer Science: Machine Learning: Structured data
Top: Computer Science: Machine Learning: Statistical Learning

You might be experiencing some problems with Your Video player.

*Slides*
0:01	Graph Mining Applications in Machine Learning Problems
2:28	Motivations for graph analysis
3:28	Graphs
4:07	Graph Structures in Biology
5:00	Overview
6:32	Path Representations & Marginalized Graph Kernels
6:35	Marginalized Graph Kernels
7:57	Label Path
9:05	Path-Probability Vector
9:48	Kernel Definition
11:37	Computation
13:35	Graph Kernel Applications
14:33	Strong Points of MGK
15:20	Drawbacks of Graph Kernels
18:06	Substructure Representation & Graph Mining
18:16	Substructure Representation
20:31	Graph Mining pt 1
21:59	Graph Mining pt 2
23:15	Enumeration on Tree-Shaped Search Space
24:52	Tree Pruning
26:30	Gspan
27:02	Depth First Search (DFS) Code
27:47	Discriminative Patterns
29:21	Multiclass Version
30:13	Summary of Graph Mining
30:58	EM-Based Clustering of Graphs (Tsudaand Kudo, ICML 2006)
31:12	EM-Based Graph Clustering
31:57	Probabilistic Model
32:45	Ordinary EM Algorithm
33:58	Regularization
35:36	E-Step
36:27	M-Step
37:28	Solution pt 1
38:18	Solution pt 2
38:22	Important Observation
38:57	Mining for Active Patterns
39:45	Experiments - RNA Graphs
41:12	Clustering RNA Graphs
42:14	Examples of RNA Graphs
42:43	ROC Scores
44:24	No of Patterns & Time
45:37	Found Patterns
46:23	Conclusion
46:58	Ongoing Work
50:20	Multiclass Version (a)

Lecture rating

People found this lecture:
Worth seeing
because it is:
Valuable and informative
Well presented
Easily understandable
Acceptably recorded
You need to login to cast your vote.

Report a problem or upload files

If you have found a problem with this lecture or would like to send us extra material, articles, exercises, etc., please use our ticket system to describe your request and upload the data.
Enter your e-mail into the 'Cc' field, and we will keep you updated with your request's status.