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

TL;DR

This paper introduces a reinforcement learning-based translation model that converts natural language expressions into search queries for conversational systems, improving query understanding and relevance feedback integration.

Contribution

It presents a novel reinforcement learning framework for NL to query translation that addresses data scarcity and enhances relevance feedback incorporation.

Findings

Effective translation from NL to queries demonstrated on TREC datasets

Reinforcement learning improves query relevance and translation accuracy

Framework overcomes dataset limitations through word selection and feedback

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 in the learning process. Experiments are carried out on two TREC datasets and outline the effectiveness of our approach.