Modeling Semantics with Gated Graph Neural Networks for Knowledge Base Question Answering

Abstract

The most approaches to Knowledge Base Question Answering are based on semantic parsing. In this paper, we address the problem of learning vector representations for complex semantic parses that consist of multiple entities and relations. Previous work largely focused on selecting the correct semantic relations for a question and disregarded the structure of the semantic parse: the connections between entities and the directions of the relations. We propose to use Gated Graph Neural Networks to encode the graph structure of the semantic parse. We show on two data sets that the graph networks outperform all baseline models that do not explicitly model the structure. The error analysis confirms that our approach can successfully process complex semantic parses.

arXiv commentAccepted as COLING 2018 Long Paper, 12 pages

References in corpus (3)

• Adam: A Method for Stochastic Optimization
• Semi-Supervised Classification with Graph Convolutional Networks
• Universal Dependencies to Logical Forms with Negation Scope

Cited by in corpus (18)

• Graph Neural Networks for Natural Language Processing: A Survey
• UHop: An Unrestricted-Hop Relation Extraction Framework for Knowledge-Based Question Answering
• Representing Schema Structure with Graph Neural Networks for Text-to-SQL Parsing
• Interpreting Graph Neural Networks for NLP With Differentiable Edge Masking
• Tripartite Heterogeneous Graph Propagation for Large-scale Social Recommendation
• A Survey on Graph Neural Networks for Knowledge Graph Completion
• Augmenting Neural Machine Translation with Knowledge Graphs
• Language Models as Knowledge Bases: On Entity Representations, Storage Capacity, and Paraphrased Queries
• Global Reasoning over Database Structures for Text-to-SQL Parsing
• Multi-Task Learning with Multi-View Attention for Answer Selection and Knowledge Base Question Answering
• Gated Graph Recursive Neural Networks for Molecular Property Prediction
• Multi-Module System for Open Domain Chinese Question Answering over Knowledge Base
• SYGMA: System for Generalizable Modular Question Answering OverKnowledge Bases
• SPARQA: Skeleton-based Semantic Parsing for Complex Questions over Knowledge Bases
• Answering Any-hop Open-domain Questions with Iterative Document Reranking
• Fea2Fea: Exploring Structural Feature Correlations via Graph Neural Networks
• Calculating Question Similarity is Enough: A New Method for KBQA Tasks
• Modeling Global Semantics for Question Answering over Knowledge Bases