Dialogue Control in the Alparon System

Rothkrantz, Léon J. M.; Vark, Robert J. van; Peters, Alexandra; Andeweg, Niels A.

doi:10.1007/3-540-45323-7_56

Cited by 4 publications

(1 citation statement)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Mellon Communicator system to model more complex tasks [10], [26], [42], [3], task structure graphs which provide a semantic structure and are used to determine the behavior of the dialogue control as well as the language understanding module [40], type hierarchies are used to model the domain of a dialogue and as a basic for clarification questions [12], blackboard is used to manage contextual information relevant to dialogue manager such as history board, control board, presentation board, etc [24].…”

Section: Information State Approachesmentioning

confidence: 99%

A modified approach of POMDP-based dialogue management

Yuan

Bian

2010

2010 IEEE International Conference on Robotics and Biomimetics

View full text Add to dashboard Cite

The author has granted a nonexclusive license allowing Library and Archives Canada to reproduce, publish, archive, preserve, conserve, communicate to the public by telecommunication or on the Internet, loan, distribute and sell theses worldwide, for commercial or noncommercial purposes, in microform, paper, electronic and/or any other formats. L'auteur a accorde une licence non exclusive permettant a la Biblioth&que et Archives Canada de reproduire, publier, archiver, sauvegarder, conserver, transmettre au public par telecommunication ou par I'lnternet, preter, distribuer et vendre des theses partout dans le monde, a des fins commerciales ou autres, sur support microforme, papier, electronique et/ou autres formats. I would like to express my gratitude to all those who gave me support in the process of my master's study. Especially, I want to thank my supervisor, Dr. Xiaobu Yuan, for his patient guidance and enthusiasm he invested in this whole research. His creativity and strong AI knowledge inspire me and light up the road of my study. Moreover, his integrity influences me and make me understand how to be a good researcher. I also want to express my appreciation as well to my thesis committee members, Dr. Wu and Dr. Rieger, who have all generously given valuable suggestions to help me improve this thesis. Also, I would like to thank all my classmates and friends who supported me in any respect during the completion of the thesis. With their kindly help and encouragement, I'm more confident in the journey of the life. Contents Author's Declaration of Originality iii Abstract iv Dedication v Acknowledgements vi List of Figures ix List of Tables x vii CONTENTS viii 5.1 POMDP Models 29 5.1.1 Flat POMDP Model 30 5.1.2 Factored POMDP Model 32 5.2 Shortcoming with the Current Models 34 5.3 A New Approach with Modification 35 6 Experiments And Discussions 6.1 Applied Domain Background 41 6.2 Implementation 6.3 Experiments and Results 7 Conclusion And Future Work Bibliography Vita Auctoris ix List of Tables 6.1 Belief States in Normal Scenario for Both Approaches 6.2 Belief States for the POMDP-based Approach 6.3 Belief States for the Proposed Approach x with the expert system to provide more friendly, flexible and accurate services to the human user to achieve particular task or provide social services. When asking for help, however, users still prefer a more natural way of communication that resembles what happens in real life, and expect more precise answers than being forced to make choices from a usually long list of possible answers. While, the real-time interaction between users and agents has created new challenges, and one of them is information exchange that ensures adequate communication for users to converse their goal of required services [16]. The media of communication normally go through the three senses of seeing, hearing, and touching, the format of communication demands for truthful, relevant, clear, and informative dialogue between users and agents. As the development of the speech recognition techniq...

show abstract

Section: Information State Approachesmentioning

confidence: 99%