0.25
0.5
0.75
1.25
1.5
1.75
2
Formal Query Generation for Question Answering over Knowledge Bases
Published on Jul 10, 2018551 Views
Question answering (QA) systems often consist of several components such as Named Entity Disambiguation (NED), Relation Extraction (RE), and Query Generation (QG). In this paper, we focus on the QG pr
Related categories
Chapter list
Formal query generation for question answering over knowledge bases00:00
Agenda00:27
Introduction - 100:57
Introduction - 201:04
Common Architectures01:34
Pipeline Architecture02:44
SPARQL Query generation (SQG)04:10
Inputs04:41
Architecture05:24
Feature identification06:03
Query Generation - Capturing subgraph07:04
Valid walks07:43
Extract valid walks from the subgraph - 108:20
Extract valid walks from the subgraph - 209:45
Ranking model09:59
LSTM10:08
Tree-LSTM10:29
Dependancy parsing tree11:08
Tree-Rep. of Candidate queries11:24
Tree-LTSM as ranking mode11:43
Evaluation setup12:29
Evaluation- Scenarios13:27
Evaluation- Ranking model14:22
Evaluation- vs. baselines15:15
Summary15:54