Markus M. Berg

An OpenCCG-Based Approach to Question Generation from Concepts

Dialogue systems are often regarded as being tedious and inflexible. We believe that one reason is rigid and inadaptable system utterances. A good dialogue system should automatically choose a formulation that reflects the user’s expectations. However, current dialogue system development environments only allow the definition of questions with unchangeable formulations. In this paper we present a new approach to the generation of system questions by only defining basic concepts. This is the basis for realising adaptive, user-tailored, and human-like system questions in dialogue systems.

Natürlichsprachlichkeit in Dialogsystemen

ZusammenfassungAktuelle Sprachdialogsysteme scheinen keinen guten Ruf zu haben. Doch woran liegt das? Welche Eigenschaften machen sie natürlich(er)? Wichtige Merkmale sind Adaptivität, gemischte Initiative, Möglichkeiten zur Korrektur und Überbeantwortung, robuste Spracherkennung, Interpretation von Umgangssprache, Verneinungen und Rückbezügen, sowie ein grundlegendes Sozialverhalten. Bei einem Vergleich der Systeme fällt jedoch auf, dass viele der Eigenschaften natürlicher Dialogsysteme noch nicht umgesetzt worden sind. Dies liegt vor allem am Fehlen eines allumfassenden Dialogmodellierungs- und -implementierungswerkzeugs. Gleichzeitig mag auch eine gewisse Diskrepanz zwischen Industrie und Wissenschaft Ursache für diesen Umstand sein. So behaupten Praktiker, dass sie natürliche Systeme gar nicht bräuchten (was eher daran liegen könnte, dass die Realisierung zu teuer bzw. schlichtweg noch nicht möglich ist). Die Wissenschaft betrachtet isolierte Probleme auf eine sehr fokussierte und detaillierte Weise, die zwar einzelne Probleme löst, jedoch aufgrund ihrer Fokussierung oftmals nicht direkt und autark in der Praxis einsetzbar ist. Hier gilt es im Rahmen der angewandten Wissenschaft gemeinsam Lösungen hinsichtlich eines Gesamtmodells zu entwickeln.

Towards interrogative types in task-oriented dialogue systems

The classification of questions and the identification of their respective answer types are crucial for different parts of a dialogue system and especially important for Rapid Application Development purposes. A common taxonomy of question types helps to connect parsers, grammars, pattern-based language generation methods and the natural language understanding module. Thus, in this paper we will present an overview of different question types and propose an abstract question description.

Integration of natural language dialogues into the conceptual model of storyboard design

Web information systems are growing with regard to complexity. Speech is a way of increasing usability and facilitating interaction. Moreover it nowadays is an important factor concerning customer needs and wishes. This paper focusses on the extension of an existing conceptual model called storyboard, to support the design of natural language dialogues. It includes a short description of storyboarding and natural language dialogues. Subsequently the outcomes are evaluated, condensed and then integrated into the storyboard model. The results of this work show that only little adaptions regarding the storyboard concept are necessary and the extension of the presentation layer with a channel-dependent renderer is sufficient to be able to model natural language dialogues.

NADIA: A Simplified Approach Towards the Development of Natural Dialogue Systems

Spoken Dialogue Systems have enormously improved during the last couple of years and gave rise to voice-controlled mobile assistants. While the abilities of these systems are very sophisticated, there is a lack of tools enabling us to easily describe a natural dialogue that can afterwards be processed by a dialogue engine without having to programme the engine itself. In this paper we present NADIA, a dialogue engine that can process an easy to define XML-based dialogue description.

Integration of dialogue patterns into the conceptual model of storyboard design

Web information systems, e.g. modern e-commerce platforms, become nowadays more sophisticated, cope with more complex applications and support an integration of speech dialogues. Their workflow and supporting infrastructure can be specified by storyboards. The integration of speech dialogues is however an unsolved issue due to the required flexibility, to the wide variety of responses and the expected nativeness. Classical keyword-based search cannot cope with such interaction media. This paper extends storyboarding by speech dialogues. Speech dialogues must be very flexible in both recognition of answers and in generation of appropriate answers. We thus introduce a pattern-based approach to specification and utilisation of speech dialogues. The paper shows that it is possible to create patterns for common dialogue-forms. Consequently they are integrated into the storyboard model and build the basis for the modeling of natural dialogues in web information systems.