Learning Automata Based Approach to Optimise Spoken Dialogue Policies

Spoken Dialogue System (SDS) is a natural language interface designed to make use of spoken language technology to accomplish a goal between the user and a computer. It also requires the development of dialogue strategies which is responsible for controlling the flow of interaction among the other components to form a complete system in an efficient and natural way according to the user behaviour and preferences. One approach to address the problem of dialogue strategy design is to use Reinforcement Learning (RL) techniques to find the solution to the problem have been a leading research area for the last few years. It is the field of research that this book seeks to propose and evaluate a scalable dialogue optimization framework.