en-de
en-es
en-fr
en-sl
en
en-zh
0.25
0.5
0.75
1.25
1.5
1.75
2
Machine Learning and Knowledge Discovery for Semantic Web
Published on Jun 02, 20116237 Views
Machine Learning and Semantic web are covering conceptually different sides of the same story - Semantic Web’s typical approach is top-down modeling of knowledge and proceeding down towards the data w
Related categories
Chapter list
Machine Learning and Knowledge Discovery for Semantic Web00:00
Jožef Stefan Institute, Artificial Intelligence Laboratory00:09
AILab Technologies02:02
Outline03:28
Motivation04:36
ML & KDD role within Semantic Web06:36
Ontology - SW commonly uses ontologies to structure knowledge 08:11
Ontology construction09:14
ML/KDD for ontology learning10:38
Ontology Population via document classification into topic ontology11:57
OntoClassify12:35
Constructing ontology from data stream15:36
Illustrative results on Reuters news17:49
Ontology Learning from text 19:13
Ontology Learning from text (cont)19:47
Semi-Automatic Data-Driven Ontology Construction20:03
Ontology Learning with OntoGen20:24
Main Features20:42
Ontology management21:18
Concept management21:58
Active Learning for concept learning22:17
Multiple views of the same data23:14
Concept’s instances visualization23:54
Ontology population24:39
OntoGen on Images28:33
Image representation29:39
Image representation - features29:46
OntoGen on ImageNet subset (flowers, fire, buildings)29:57
Document list for quick overview30:24
Collection visualization (without displaying images)30:30
Collection visualization (displaying images)30:51
Creating ontology on images31:04
Sub-concept visualization31:39
Adding sub-concepts31:56
Text-Driven Ontology extension32:16
OntoPlus (1)32:33
OntoPlus (2)34:40
Results – Concept Ranking35:25
Demo 37:20
Context Sensitive Search38:14
SearchPoint38:42
Approach Description39:09
Search39:25
Example – Searching for “jaguar”39:29
Context sensitive search39:50
SearchPoint (1)40:51
SearchPoint (2)40:55
Main advantages41:23
SearchPoint integrated in Accenture’s intranet search41:24
Answer Art46:53
AnswerArt – System Architecture47:26
AnswerArt using Medline (1)48:40
AnswerArt using Medline (2)48:59
AnswerArt using Medline (3)49:50
AnswerArt using Medline (4)50:04
AnswerArt using ASFA (1)50:32
AnswerArt using ASFA (2)50:39
AnswerArt using ASFA (3)50:54
Natural Language Text Enrichment51:05
Enrycher Service53:51
Entity resolution in text54:55
Enrycher Service Dependencies56:04
A comparative view on five systems: Enrycher, Text Runner, Open Calais, GATE and Read the Web56:32
Enrycher - demo (1)56:46
Enrycher - demo (2)56:51
Enrycher - demo (3)57:00
Enrycher - demo (4)58:11
Opinion Mining 59:59
Opinion Mining - Use case: Twitter comments on movies 01:00:29
Twitter comments analysis01:02:09
Sensor Search01:03:03
Sensor Search - Architecture01:03:17
Sensor Search01:04:10
Real -Time information processing01:06:20
QMiner01:06:38
Network Monitoring for British Telecom01:08:03
Visualizing Root-cause and prediction01:09:06
How Well Are We Predicting01:09:27
User Modeling for NYTimes & Bloomberg01:10:10
Generalizing from registered users01:11:23
Using User Modeling for News Recommendations01:12:24
Recommendation01:12:48
Real-time Architecture01:14:15
Results01:14:51
User Recommendation01:16:19
Personalized News Search01:17:21
Personalized News Search – System Architecture01:17:22
User: Young female computer programmer / Query: Religion01:17:23
User: Middle aged male clergy / Query: Religion01:17:51
Videolectures.net01:17:59
Montreal @ Video Lectures01:18:13