en-es
en
0.25
0.5
0.75
1.25
1.5
1.75
2
Knowledge Graph Identification
Published on Nov 28, 20135263 Views
Large-scale information processing systems are able to extract massive collections of interrelated facts, but unfortunately transforming these candidate facts into useful knowledge is a formidable cha
Related categories
Chapter list
Knowledge Graph Identification00:00
Overview00:04
Challenges in Knowledge Graph Construction00:47
Motivating Problem: New Opportunities00:50
Motivating Problem: Real Challenges01:11
NELL:The Never-Ending Language Learner01:35
Examples of NELL errors02:15
Entity co-reference errors02:17
Missing and spurious labels02:37
Missing and spurious relations02:55
Violations of ontological knowledge03:12
Knowledge Graph Identification - 103:51
Motivating Problem (revised)03:57
Knowledge Graph Identification - 204:17
Illustration of KGI: Extractions 04:44
Illustration of KGI: Extraction Graph04:54
Illustration of KGI: Ontology + ER05:05
Illustration of KGI05:16
Modeling Knowledge Graph Identification05:36
Viewing KGI as a probabilistic graphical model - 105:38
Viewing KGI as a probabilistic graphical model - 205:56
Background: Probabilistic Soft Logic (PSL)06:26
Background: PSL Rules to Distributions07:04
Background: Finding the best knowledge graph07:46
PSL Rules for the KGI Model08:07
PSL Rules: Uncertain Extractions - 108:10
PSL Rules: Uncertain Extractions - 208:13
PSL Rules: Uncertain Extractions - 308:18
PSL Rules: Uncertain Extractions - 408:32
PSL Rules: Entity Resolution08:37
PSL Rules: Ontology08:53
Evaluation09:12
Two Evaluation Datasets09:14
LinkedBrainz dataset for KGI09:50
Adding noise to LinkedBrainz10:10
LinkedBrainzexperiments10:45
NELL Evaluation: two settings11:18
NELL experiments: Target Set12:15
NELL experiments: Complete knowledge graph13:13
Conclusion13:32