Proceedings of the Symposium on Dialogue Modelling and Generation

Permanent URI for this collection

http://hdl.handle.net/2003/21460

Organizers and Editors:

Sassen, Claudia
Piwek, Paul
Kühnlein, Peter

Program Committee

Anton Benz, Syddansk Univ., Denmark
Harry Bunt, Tilburg Univ., Netherlands
Simon Garrod, Univ. of Glasgow, UK
Emiel Krahmer, Tilburg University, Netherlands
Ivana Kruijff-Korbayová, Univ. des Saarlandes, Germany
Peter Kühnlein, Univ. of Bielefeld, Germany
Colin Matheson, Univ. of Edinburgh, UK
Daniel Paiva, Univ. of Sussex, UK
Paul Piwek, Open University, UK
Richard Power, ITRI, Univ. of Brighton, UK
Matthew Purver, CSLI, USA
Alison Sanford, Univ. of Strathclyde, UK
Claudia Sassen, Univ. of Dortmund, Germany
Donia Scott, ITRI, Univ. of Brighton, UK
Marilyn Walker, Univ. of Sheffield, UK

Wird unter der ISBN 3-921823-30-7 erscheinen. Dezember 2007: die Proceedings sind jetzt als Sonderband 44(3) /Dialogue Modelling and Generation/ der Zeitschrift "DIscourse Processes" im Verlag Lawrence Erlbaum erschienen. Die Guest Editors für diesen Band sind Peter Kühnlein und Paul Piwek.

Browse

Now showing 1 - 8 of 8

Conceptual Mismatches and Repair in Human-Computer Interaction
(University of Dortmund, 2005-06-13) Beun, Robbert-Jan; van Eijk, Rogier M.
We present a computational framework for the generation of elementary speech acts to establish conceptual alignment between a computer system and its user. We clearly distinguish between two phases of the alignment process: message interpretation and message generation. In the interpretation phase, presuppositions are extracted from the user’s message and compared with the system’s ontology. Subsequently, in the generation phase, an adequate feedback message is produced in order to resolve detected discrepancies. We provide a conversational strategy that is based upon Gricean implicatures and a distinction between three types of beliefs: private beliefs about the domain of discourse, beliefs about the beliefs of the other and beliefs about the shared beliefs.
Dialogue Modeling in Embodied Communicational Agents
(Universität Dortmund, 2005-07-01) Laufer, László; Tatai, Gábor
We are developing an ECA (Embodied Communicational Agent) system in Hungarian language. The ECA has to be able to carry out general chat conversations and domain specific discussions as well. The structure of the dialogue graph and the attributes of the graph nodes have to support the success of the conversation: user utterance understanding and responding, as well as repairing the 'non-understandins' situations. In the following sections we are briefly depicting the system and describe its dialogue knowledge base focusing on the different user interfaces we developed for its testing and building.
Modeling Speech Disfluency to Predict Conceptual Misalignment in Speech Survey Interfaces
(University of Dortmund, 2005-06-13) Ehlen, Patrick; Schober, Michael F.
Computer-based interviewing systems could use models of respondent disfluency behaviors to predict a need for clarification of terms in survey questions. We compare simulated speech interfaces that use two such models - a generic model and a stereotyped model that distinguishes between the speech of younger and older speakers - to several non-modeling speech interfaces in a task where respondents provided answers to survey questions from fictional scenarios. Our modeling procedure found that the best predictor of conceptual misalignment was a critical Goldilocks range for response latency, outside of which responses are more likely to be conceptually misaligned. Different Goldilocks ranges are effective for younger and older speakers.
Towards the Generation of Overspecified Multimodal Referring Expressions
(University of Dortmund, 2005-06-13) Krahmer, Emiel; van der Sluis, Ielka
Generating Individualized Utterances for Dialogue Systems
(University of Dortmund, 2005-06-13) Mairesse, Francois; Walker, Marlyn
One of the most robust findings of studies of human-human dialogue is that people adapt their utterances to their conversational partners. However, spoken language generators are limited in their ability to adapt to individual users. While statistical models of language generation have the potential for individual adaptation, we know of no experiments showing this. In this paper, we utilize one statistical method, boosting, to train a spoken language generator for individual users. We show that individualized models perform better than models based on sets of users, and describe differences in the learned individual models arising from the linguistic preferences of users.
On the Role of Conceptual and Linguistic Ontologies in Spoken Dialogue Systems
(University of Dortmund, 2005-06-13) Bateman, John; Farrar, Scott; Ross, Robert J.; Tenbrink, Thora
We report on the role of well-formed conceptual and linguistic ontologies in empirically grounded `spoken dialogue systems' (SDS). In particular we use empirical results from spatial dialogues in German to argue for the strict separation of linguistically motivated knowledge from non-linguistic, domain concerns. We motivate our arguments with a number of examples relevant to the language generation task, and show how a well-defined separation of linguistic and domain concerns can be effected in a practical SDS.
Developing Human-Robot Dialogue Management Formally
(University of Dortmund, 2005-06-13) Bateman, John; Shi, Hui
In shared-control systems, such as intelligent service robots, a human operator and an automated technical system are interdependently in charge of control. Natural Language dialogues have long been acknowledged as a potentially fruitful modality for instructing, describing and negotiating in human-machine interfaces. Since shared-control systems are often embedded in safety-critical devices, formal methods are thus widely used for improving the quality of such systems. In this paper, we present a formal method based approach for dialogue management and show how it enhances the clarity of dialog modelling, provides several engineering properties (e.g., validation, test and simulation) and supports the generation of clarification subdialogues.
Using Generalised Dialogue Models to Constrain Information State Based Dialogue Systems
(University of Dortmund, 2005-06-13) Bateman, John; Ross, Robert J.; Shi, Hui
While Information State (IS) based techniques show promise in the construction of flexible, knowledgebased dialogue systems, the many declarative rules that are used to encode Dialogue Theories often lead to opaque systems that are difficult to test and potentially unintuitive to users. In this paper, we advocate the application of explicitly defined Generic Dialogue Models (GDMs), encoded as recursive transition networks (RTNs), to the structuring of information state-based dialogue managers. To this end, we review the state of GDM approaches, comparing and contrasting them against the Dialogue Theories which are typically implemented using information state approaches. Furthermore, to support our approach, we present an extension of the ALPHA (A Language for Programming Hybrid Agents) language, which has been enhanced to support information state and GDM concepts directly.

Browse

Recent Submissions