Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Conversational question answering over knowledge graphs with transformer and graph attention networks

: Kacupaj, E.; Plepi, J.; Singh, K.; Thakkar, H.; Lehmann, J.; Maleshkova, M.

Volltext ()

Merlo, P. ; Association for Computational Linguistics -ACL-, European Chapter -EACL-:
16th Conference of the European Chapter of the Associationfor Computational Linguistics, EACL 2021. Online resource : Proceedings of the conference : April 19-23, 2021
Stroudsburg, PA: ACL, 2021
ISBN: 978-1-954085-02-2
Association for Computational Linguistics, European Chapter (EACL Conference) <16, 2021, Online>
Konferenzbeitrag, Elektronische Publikation
Fraunhofer IAIS ()

This paper addresses the task of (complex) conversational question answering over a knowledge graph. For this task, we propose LASAGNE (muLti-task semAntic parSing with trAnsformer and Graph atteNtion nEtworks). It is the first approach, which employs a transformer architecture extended with Graph Attention Networks for multi-task neural semantic parsing. LASAGNE uses a transformer model for generating the base logical forms, while the Graph Attention model is used to exploit correlations between (entity) types and predicates to produce node representations. LASAGNE also includes a novel entity recognition module which detects, links, and ranks all relevant entities in the question context. We evaluate LASAGNE on a standard dataset for complex sequential question answering, on which it outperforms existing baseline averages on all question types. Specifically, we show that LASAGNE improves the F1-score on eight out of ten question types; in some cases, the increase in F 1-score is more than 20% compared to the state of the art.