Hakimov, Sherzod: Learning Multilingual Semantic Parsers for Question Answering over Linked Data. A comparison of neural and probabilistic graphical model architectures. 2019
Inhalt
- Abstract
- Acknowledgements
- Introduction
- Natural Language Interfaces
- Semantic Parsing
- Natural Language Question Answering System
- Motivation
- Research Questions
- Contributions
- Chapter Structure
- Preliminaries
- RDF & Semantic Web
- Syntax
- Semantics
- Factor Graphs
- Neural Networks
- Datasets
- Related Work
- Lexicon
- CCG-based Semantic Parsing Approach
- Dependency parse tree-based Semantic Parsing Approach
- Overview
- Dependency Parse Tree-based Approach
- Learning Model Parameters
- Evaluation
- Discussion
- Neural Network-based Semantic Parsing Approach
- Overview
- Methods
- Inverted Index Construction for Entity Retrieval
- Named Entity Recognition
- Candidate Pair Generation
- Model 1: BiLSTM-Softmax
- Model 2: BiLSTM-KB
- Model 3: BiLSTM-Binary
- Model 4: FastText-Softmax
- Evaluation
- Discussion
- Discussion
- Overview
- Manual Effort
- CCG-based Semantic Parsing Approach
- Dependency parse tree-based Semantic Parsing Approach
- Neural Network-based Semantic Parsing Approach
- Comparison
- Syntax and Semantics Relationship
- CCG-based Semantic Parsing Approach
- Dependency parse tree-based Semantic Parsing Approach
- Neural Network-based Semantic Parsing Approach
- Comparison
- Multilinguality
- CCG-based Semantic Parsing Approach
- Dependency parse tree-based Semantic Parsing Approach
- Neural Network-based Semantic Parsing Approach
- Comparison
- Cross-domain Transferability
- CCG-based Semantic Parsing Approach
- Dependency parse tree-based Semantic Parsing Approach
- Neural Network-based Semantic Parsing Approach
- Comparison
- Training Data Size & Search Space
- Conclusion
- QALD Dataset
- Bibliography