Abstract | ||
---|---|---|
The overall objective of u0027socialu0027 dialogue systems is to support engaging, entertaining, and lengthy conversations on a wide variety of topics, including social chit-chat. Apart from raw dialogue data, user-provided ratings are the most common signal used to train such systems to produce engaging responses. In this paper we show that social dialogue systems can be trained effectively from raw unannotated data. Using a dataset of real conversations collected in the 2017 Alexa Prize challenge, we developed a neural ranker for selecting u0027goodu0027 system responses to user utterances, i.e. responses which are likely to lead to long and engaging conversations. We show that (1) our neural ranker consistently outperforms several strong baselines when trained to optimise for user ratings; (2) when trained on larger amounts of data and only using conversation length as the objective, the ranker performs better than the one trained using ratings -- ultimately reaching a Precision@1 of 0.87. This advance will make data collection for social conversational agents simpler and less expensive in the future. |
Year | DOI | Venue |
---|---|---|
2018 | 10.18653/v1/w18-5701 | SCAI@EMNLP |
DocType | Volume | ISSN |
Journal | abs/1811.00967 | Proceedings of the 2018 EMNLP Workshop SCAI: The 2nd International
Workshop on Search-Oriented Conversational AI, pages 1-8. ISBN
978-1-948087-75-9 |
Citations | PageRank | References |
3 | 0.38 | 17 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Igor Shalyminov | 1 | 9 | 3.30 |
Ondřej Dušek | 2 | 180 | 23.08 |
Oliver Lemon | 3 | 135 | 14.94 |