DESIGN PRINCIPLES OF THE DISCO

SYSTEM

Gunter Neumann?

Deutsches Forschungszentrum fur Kunstliche Intelligenz

Stuhlsatzenhausweg 3

D-6600 Saarbrucken 11, Germany

[email protected]
To appear in: Proceedings of the TWLT 5, Twente, Netherlands, 1993.

abstract

In this paper we introduce the basic design principles of the disco system, a Natural Language analysis and generation system. In particular we describe the disco development shell, the basic tool for the integration of natural language components in the disco system, and its application in the cosma (COoperative Schedule MAnagement Agent) system.

Following an object oriented architectural model we introduce a two{step approach, where in the first phase the architecture is developed independently of specific components to be used and of a particular flow of control. In the second phase the frame system" is instantiated by the integration of existing components as well as by defining the particular flow of control between these components. Because of the object{oriented paradigm it is easy to augment the frame system, which increases the flexibility of the whole system

?The research underlying this paper was supported by a research grant, FKZ ITW 9002 0, from the German Bundesministerium fur Forschung und Technologie to the DFKI project disco.

with respect to new applications. The development of the cosma system will serve as an example of this claim.

1 introduction

Today's natural language systems are large software products. They consist of serveral mutually connected components of different kinds, each developed by different researchers often placed on different locations. The integration of these components has therefore become a software engineering and mangement problem. We will consider the project disco (DIalogue Systems for COoperating agents) from this perspective.
disco's primary goal is the processing of multiagent natural language dialogue. Multiagent capabilities make it an appropriate front end for autonomous cooperative agents, which will be exemplified by the cosma system (COoperative Scheduling MAnagement system) described in section 4.

The project disco as a whole is a fouryear, eight-person research effort funded by the German Ministry for Research and Technology. disco is completing its