Implementation of a Statistical Dialogue Manager for Commercial Conversational Systems

Abstract

Conversational interfaces have recently become an ubiquitous element in both the personal sphere by improving individual’s quality of life, and industrial environments by the automation of services and its corresponding costs savings. However, designing the dialogue model used by these interfaces to decide the next response is a hard-to-accomplish task for complex conversational interactions. In this paper, we propose a statistical-based dialogue manager architecture, which provides flexibility to develop and maintain this module. Our proposal has been integrated with DialogFlow, a natural language understanding platform provided by Google to design conversational user interfaces. The proposed architecture has been assessed with a real use case for a train scheduling domain, proving that the user experience is of a high value and it can be integrated for commercial setups.