AMUSE: Multilingual Semantic Parsing for Question Answering over Linked Data thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

AMUSE: Multilingual Semantic Parsing for Question Answering over Linked Data

Published on Nov 28, 20171160 Views

The task of answering natural language questions over RDF data has received wIde interest in recent years, in particular in the context of the series of QALD benchmarks. The task consists of mapping a

Related categories

Chapter list

AMUSE: Multilingual Semantic Parsing for Question Answering over Linked Data00:00
Virtual Assistants00:12
Multilingual - Factoid Question Answering00:36
Problem Definition - Question Answering01:29
Motivation03:02
Knowledge Base - DBpedia04:28
Preliminaries05:03
Logical Form - DUDES05:26
DUDES06:08
Semantic Composition with DUDES07:31
Semantic Composition with DUDES - Bottom Up08:13
Semantic Composition with DUDES - Bottom Up - 109:00
Semantic Composition with DUDES - Bottom Up - 209:15
Semantic Composition with DUDES - 109:40
Logical Form into SPARQL09:55
Factor Graphs10:19
Approach11:11
Approach - 111:23
Inference11:43
Inference - 112:04
Linking to Knowledge Base (L2KB)12:34
Linking to Knowledge Base (L2KB) - 113:15
Query Construction (QC)13:39
Query Construction (QC) - 113:57
Features14:17
Addressing The Lexical Gap14:28
Addressing The Lexical Gap - 114:28
Evaluation - Lexicon14:36
Evaluation - Question Answering15:06
Evaluation - Question Answering - 115:32
System Errors16:20
Conclusion16:35
Future directions16:48
Thanks!17:17