Semantic Parsing for Conversational Question Answering over Knowledge Graphs
Preprint
Abstract
In this paper, we are interested in developing semantic parsers which understand natural language questions embedded in a conversation with a user and ground them to formal queries over definitions in a general purpose knowledge graph (KG) with very large vocabularies (covering thousands of concept names and relations, and millions of entities). To this end, we develop a dataset where user questions are annotated with Sparql parses and system answers correspond to execution results thereof. We present two different semantic parsing approaches and highlight the challenges of the task: dealing with large vocabularies, modelling conversation context, predicting queries with multiple entities, and generalising to new questions at test time. We hope our dataset will serve as useful testbed for the development of conversational semantic parsers. Our dataset and models are released at https://github.com/EdinburghNLP/SPICE.Keywords
All Related Versions
This publication has 22 references indexed in Scilit:
- Memory-Based Semantic ParsingTransactions of the Association for Computational Linguistics, 2021
- A Survey on Conversational Recommender SystemsACM Computing Surveys, 2021
- Beyond I.I.D.: Three Levels of Generalization for Question Answering on Knowledge Bases2021
- Structured Context and High-Coverage Grammar for Conversational Question Answering over Knowledge Graphs2021
- Conversational Question Answering over Knowledge Graphs with Transformer and Graph Attention Networks2021
- On Compositional Generalization of Neural Machine Translation2021
- COGS: A Compositional Generalization Challenge Based on Semantic Interpretation2020
- A Theoretical Framework for Conversational Search2017
- Large-scale Semantic Parsing without Question-Answer PairsTransactions of the Association for Computational Linguistics, 2014
- Wikidata2012