A Reinforcement Learning-driven Translation Model for Search-Oriented Conversational Systems

Abstract : Search-oriented conversational systems rely on information needs expressed in natural language (NL). We focus here on the understanding of NL expressions for building keyword-based queries. We propose a reinforcement-learning-driven translation model framework able to 1) learn the translation from NL expressions to queries in a supervised way, and, 2) to overcome the lack of large-scale dataset by framing the translation model as a word selection approach and injecting relevance feedback as a reward in the learning process. Experiments are carried out on two TREC datasets. We outline the effectiveness of our approach.
Complete list of metadatas

Cited literature [20 references]  Display  Hide  Download

https://hal.sorbonne-universite.fr/hal-02002991
Contributor : Laure Soulier <>
Submitted on : Friday, June 7, 2019 - 12:45:30 PM
Last modification on : Friday, July 5, 2019 - 3:26:03 PM

File

SCAI_2018___RL_translation.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02002991, version 1

Citation

Wafa Aissa, Laure Soulier, Ludovic Denoyer. A Reinforcement Learning-driven Translation Model for Search-Oriented Conversational Systems. SCAI 2018 : 2nd edition of the workshop on Search-Oriented Conversational AI, Oct 2018, Bruxelles, Belgium. pp.33 - 39. ⟨hal-02002991⟩

Share

Metrics

Record views

29

Files downloads

10