en
0.25
0.5
0.75
1.25
1.5
1.75
2
Constructing a Focused Taxonomy from a Document Collection
Published on Jul 08, 20133819 Views
We describe a new method for constructing custom taxonomies from document collections. It involves identifying relevant concepts and entities in text; linking them to knowledge sources like Wikipedia,
Related categories
Chapter list
Constructing a Focused Taxonomy from a Document Collection00:00
Why?00:43
Talk Overview01:27
Taxonomy Generation Research Team01:45
How Taxonomy Generation Works02:04
Taxonomy Generation Overview02:07
Taxonomy Generation -Detailed02:27
Taxonomy Generation in 5 Steps!02:30
Step 1. Document input & conversion03:04
Step 2. Extracting concepts03:24
Step 3. Annotation with meaning - 104:16
Step 3. Annotation with meaning - 204:35
Step 4. Discarding irrelevant meanings04:51
Step 5a. Group taxonomy05:44
Step 5b. Consolidating taxonomy06:19
The RDF data model07:25
Taxonomy Generation Process08:02
How Does It Look Like?09:01
Case Study & Evaluation: A News Group09:04
Case Study: A News Group - 109:44
Evaluation10:20
Evaluation: Sources of error in concept identification12:12
Case Study: A News Group - 213:22
Case Study: A News Group - 313:50
Case Study: A News Group - 414:08
Alternative Labels - 114:11
Alternative Labels - 214:30
Labels & Relations14:45
Case Study: A News Group - 515:01
Case Study: A News Group - 617:30
Case Study: A News Group - 717:36
Case Study: A News Group - 817:39
Case Study: A News Group - 917:47
Other Use Cases18:22
Other Use Cases: Discovery19:09
Summary20:38
More21:05