Neural Machine Translation for Query Construction and Composition

Abstract

Research on question answering with knowledge base has recently seen an increasing use of deep architectures. In this extended abstract, we study the application of the neural machine translation paradigm for question parsing. We employ a sequence-to-sequence model to learn graph patterns in the SPARQL graph query language and their compositions. Instead of inducing the programs through question-answer pairs, we expect a semi-supervised approach, where alignments between questions and queries are built through templates. We argue that the coverage of language utterances can be expanded using late notable works in natural language generation.

arXiv commentICML workshop on Neural Abstract Machines & Program Induction v2 (NAMPI), extended abstract

References in corpus (5)

• Universal Sentence Encoder
• Seq2SQL: Generating Structured Queries from Natural Language using Reinforcement Learning
• Neural Programmer-Interpreters
• Neural Symbolic Machines: Learning Semantic Parsers on Freebase with Weak Supervision
• SPARQL as a Foreign Language

Cited by in corpus (5)

• Knowledge Graph Question Answering using Graph-Pattern Isomorphism
• Neural Machine Translating from Natural Language to SPARQL
• Towards Natural Language Question Answering over Earth Observation Linked Data using Attention-based Neural Machine Translation
• Would You Ask it that Way? Measuring and Improving Question Naturalness for Knowledge Graph Question Answering
• SPARKLE: Enhancing SPARQL Generation with Direct KG Integration in Decoding