Dialogue System Augmented with Commonsense Knowledge
Ilya Lasy
Vilnius University, Lithuania
Virginijus Marcinkevičius
Vilnius University, Lithuania
Published 2022-05-13


Natural Language Generation
Dialogue System
Knowledge Graph
Deep Learning

How to Cite

Lasy I. and Marcinkevičius V. (2022) “Dialogue System Augmented with Commonsense Knowledge”, Vilnius University Open Series, pp. 68-76. doi: 10.15388/LMITT.2022.7.


Building an open-domain dialog system is a challenging task in current research. In order to successfully maintain a conversation with human, a dialog system must develop many qualities: being engaging, empathetic, show a unique personality and having general knowledge about the world. Prior research has shown that it is possible to develop such chat-bot system that combines these features, but this work explores this problem further. Most state-of-theart dialogue systems are guided by unstructured knowledge such as Wikipedia articles, but there is a lack of research on how structured knowledge bases can be used for open-domain dialogue generation. This work proposes usage of structured knowledge base ConceptNet for knowledge-grounded dialogue generation. Novel knowledge extraction algorithm is proposed which is then used to incorporate knowledge into existing dialogue datasets. Current state-of-theart model BlenderBot is finetuned on new datasets which shows improvement in novelty of utterances generated by the model.

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.