Sign inSign up
A Graph-Based Approach to Learn Semantic Descriptions of Data Sources thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

A Graph-Based Approach to Learn Semantic Descriptions of Data Sources

Published on Nov 28, 20134712 Views

Craig A. Knoblock

Semantic models of data sources and services provide support to automate many tasks such as source discovery, data integration, and service composition, but writing these semantic descriptions by hand

ISWC 2013 - Sydney

Related categories

Information DesignSemantic Web

Chapter list

A Graph-based Approach to Learn Semantic Descriptions of Data Sources00:00
Problem: How to learn semantic descriptions?00:09
First, what is a semantic description?00:16
Semantic Description00:22
Semantic Types01:11
Relationships01:46
Previous approach to learn semantic descriptions02:25
Karma02:26
Refining The Model - 103:35
Refining The Model - 203:56
Refining The Model - 304:08
Our new approach to learn semantic descriptions04:24
Key idea04:37
Approach05:07
Example05:55
Build a Graph from Known Models - 107:00
Build a Graph from Known Models - 208:15
Build a Graph from Known Models - 308:44
Learn Semantic Types (Previous Work)10:37
Generate Candidate Models - 111:02
Generate Candidate Models - 211:53
Generate Candidate Models - 312:11
Generate Candidate Models - 412:28
Generate Candidate Models - 512:35
Generate Candidate Models - 612:47
Generate Candidate Models - 712:48
Generate Candidate Models - 812:56
Generate Candidate Models - 912:59
Rank Source Models13:18
Evaluation13:36
Results - Dataset 114:41
Results - Dataset 215:18
Related Work15:31
Discussion16:37
Future Work17:12