论文信息 - Probabilistic Dialog Management

Probabilistic Dialog Management

Modeling user interfaces as dialogs provides a conceptual framework to address global coherence and efficiency of interactions. While non-probabilistic approaches provide convincing results and transparent dialog behavior, probabilistic techniques can help to account for inherent uncertainties in user input. In this paper, we present three patterns for probabilistic dialog management or support thereof.

[1] Dirk Schnelle-Walka,et al. A Pattern Language for Dialogue Management , 2012 .

[2] Steve J. Young,et al. Partially observable Markov decision processes for spoken dialog systems , 2007, Comput. Speech Lang..

[3] Yuan,et al. Trainbot: A spoken dialog sytem using partially observable Markov Decision Processes , 2010 .

[4] Gerard G. Meszaros,et al. A pattern language for pattern writing , 1997 .

[5] Jakob Nielsen,et al. Classification of dialog techniques , 1987, SGCH.

[6] Melanie Hartmann,et al. Prediction Algorithms for User Actions , 2007, LWA.

[7] Steve J. Young,et al. USING POMDPS FOR DIALOG MANAGEMENT , 2006, 2006 IEEE Spoken Language Technology Workshop.

[8] James O. Coplien,et al. A generative development-process pattern language , 1995 .

[9] Steve Young,et al. Statistical methods for building robust spoken dialogue systems in an automobile , 2012 .