On-Line Learning of a Persian Spoken Dialogue System Using Real Training Data

Journal of Advances in Computer Research

مقاله 4، دوره 1، شماره 2، بهمن 2010، صفحه 31-39   

چکیده

The first spoken dialogue system developed for the Persian language is introduced. This is a ticket reservation system with Persian ASR and NLU modules. The focus of the paper is on learning the dialogue management module. In this work, real on-line training data are used during the learning process. For on-line learning, the effect of the variations of discount factor (g) on the learning speed is investigated as the second contribution of the research. The optimal values for g were found and the variation pattern of the action-value function (Q) in the learning process was obtained. A probabilistic policy for selecting actions is used in this work for the first time instead of greedy policies employed in previous works.

کلیدواژه‌ها

Learning spoken dialogue؛ action value function؛ discount factor