International Conference on Machine Learning - Bonn 2005

Statistical Relational Learning - Part 3

author: Lise Getoor, University of Maryland

Description

Statistical relational learning raises many new challenges and opportunities. Because the statistical model depends on the domain's relational structure, parameters in the model are often tied. This has advantages for making parameter estimation feasible, but complicates the model search. Because the "features" involve relationships among multiple objects, there is often a need to intelligently construct aggregates and other relational features.

You might be experiencing some problems with Your Video player.

*Slides*
0:00	Four SRL Approaches
0:36	Markov Networks
2:46	Markov Network Semantics
3:55	Four SRL Approaches
4:22	Advantages of Undirected Models
5:06	Relational Markov Networks
6:32	RMNs
8:17	Collective Classification Overview
8:58	Learning RMNs
9:06	Learning RMNs
9:13	Four SRL Approaches
9:20	Markov Logic Networks
10:16	Example of an MLN
10:50	Example of an MLN
10:56	Example of an MLN
11:25	Example of an MLN
11:35	Example of an MLN
12:10	Markov Logic Networks
12:39	Summary: Undirected Approaches
13:56	Four SRL Approaches
14:20	Themes: Representation
14:51	Themes: Representational Issues
15:25	Themes: Inference
15:31	Themes: Learning
15:37	Goals
16:31	Roadmap
16:34	Discriminative Probabilistic Models for Relational Data
16:44	WebKB: Standard Classification
17:44	WebKB: Standard Classification
17:58	WebKB: Collective Classification
18:10	WebKB: Collective Classification
18:24	More Complex Structure
18:28	More Complex Structure
18:34	Collective Classification: Results
19:08	View Learning: An Extension to SRL With an Application in Mammography
19:40	Application
19:56	View Learning
20:16	Key New Predicate I
20:20	Key New Predicate II
20:33	slide42
21:12	NASD Stock Fraud Example
21:23	NASD Data
21:38	NASD Query
21:49	Predicting Ground Truth
22:19	Link-based Classification
22:29	Link-based Classification
22:46	Our Approach
22:55	Experiment
23:36	Generalized Entity Resolution
23:48	Entity Resolution
24:25	Simple Example: Author Resolution
25:52	Evaluation Datasets
25:58	Results: Best F1
26:33	Goals
26:54	Conclusion
27:39	Thanks!!!

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.