Neural Machine Translation for Query Construction and Composition
This work addresses query construction for knowledge base question answering, but it appears incremental as it builds on existing neural machine translation and natural language generation methods.
The paper tackles the problem of question parsing for knowledge base question answering by applying neural machine translation to learn SPARQL graph query patterns and their compositions, using a semi-supervised approach with templates instead of question-answer pairs.
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.