A Reinforcement Learning-driven Translation Model for Search-Oriented Conversational Systems
This addresses the challenge of understanding user queries in conversational search systems, but appears incremental as it builds on existing translation and reinforcement learning methods.
The paper tackles the problem of translating natural language expressions into keyword-based queries for search-oriented conversational systems, proposing a reinforcement learning-driven translation model that learns supervised translation and overcomes dataset limitations through word selection and relevance feedback, with experiments on TREC datasets showing effectiveness.
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 in the learning process. Experiments are carried out on two TREC datasets and outline the effectiveness of our approach.