Q. Can Knowledge Graphs be used to Answer Boolean Questions? A. It's complicated!
Citation:
Dzendzik, D., Vogel, C., Foster, J., Q. Can Knowledge Graphs be used to Answer Boolean Questions? A. It's complicated!, Proceedings of the First Workshop on Insights from Negative Results in NLP, Association for Computational Linguistics, 2020, 6-14Download Item:
Abstract:
In this paper we explore the problem of machine reading comprehension, focusing on the BoolQ dataset of Yes/No questions. We carryout an error analysis of a BERT-based machine reading comprehension model on this dataset, revealing issues such as unstable model behaviour and some noise within the dataset itself. We then experiment with two approaches for integrating information from knowledge graphs: (i) concatenating knowledge graph triples to text passages and (ii) encoding knowledge with a Graph Neural Network. Neither of these approaches show a clear improvement and we hypothesize that this may be due to a combination of inaccuracies in the knowledge graph, imprecision in entity linking, and the models’ inability to capture additional information from knowledge graphs.
Sponsor
Grant Number
Science Foundation Ireland (SFI)
13/RC/2106
Author's Homepage:
http://people.tcd.ie/vogelDescription:
PUBLISHED
Author: Vogel, Carl
Other Titles:
First Workshop on Insights from Negative Results in NLPPublisher:
Association for Computational LinguisticsType of material:
Conference PaperCollections
Availability:
Full text availableKeywords:
Model behaviour, Knowledge graphs, Graph Neural Network, BoolQ dataset, BERT-based machine reading comprehension modelSubject (TCD):
Digital Engagement , Computational Linguistics , Computational linguistics , MACHINE LEARNING , Machine Learning in Mulitmedia Information Retrieval , computational linguistics , signal processing and machine learningDOI:
https://doi.org/10.18653/v1/2020.insights-1.2Metadata
Show full item recordLicences: