Abstract
Interactive Restricted Domain Question Answering Systems combine the interactivity of dialogue systems with the information retrieval features of question answering systems. The main problem when going from task-oriented dialogue systems to interactive restricted domain question answering systems is that the lack of task structure prohibits making simplifying assumptions as in taskoriented dialogue systems. In order to address this issue, we propose a solution that combines representations based on keywords extracted from the user utterances with machine learning to learn the dialogue management function. More specifically, we propose to use Support Vector Machines to classify the dialogue state containing the extracted keywords in order to determine the next action to be taken by the dialogue manager. Much of the content selection for clarification question usually found in dialogue managers is moved to an instance-based generation component. The proposed method has the advantage that it does not rely on an explicit representation of task structure as is necessary for task-oriented dialogue systems.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2008 Springer Science + Business Media B.V.
About this chapter
Cite this chapter
Denecke, M., Yasuda, N. (2008). Does This Answer Your Question?. In: Dybkjær, L., Minker, W. (eds) Recent Trends in Discourse and Dialogue. Text, Speech and Language Technology, vol 39. Springer, Dordrecht. https://doi.org/10.1007/978-1-4020-6821-8_9
Download citation
DOI: https://doi.org/10.1007/978-1-4020-6821-8_9
Publisher Name: Springer, Dordrecht
Print ISBN: 978-1-4020-6820-1
Online ISBN: 978-1-4020-6821-8
eBook Packages: Humanities, Social Sciences and LawSocial Sciences (R0)