THE ANATOMY OF THE NATURAL LANGUAGE DIALOGUE SYSTEM HAM-RPM

Abstract

HAM-RPM is a dialogue system which converses with a human partner in colloquial German about limited, but interchangeable scenes. The objective of this report is to give a detailed, complete and self-contained description of the system in its present state of implementation. After a discussion of the goals and methodological principles which guide our research and a short introduction to the implementation language an overview of the system's architecture, of its knowledge base and of the domains of discourse is given. Then each processing phase from the analysis of natural-language input to the generation of a natural-language utterance is described in detail. The examples used during these descriptions are supplemented by transcripts of complete dialogue sessions. Finally HAM-RPM's programming environment is described. 1 THEORETICAL AND TECHNICAL BACKGROUND OF HAM-RPM 1.1 UNDERLYING ISSUES HAM-RPM is a dialogue system which converses with a human partner in colloquial German about limited, but interchangeable scenes. Recently, much AI research even in non-linguistic areas has been seeking contact with natural-language simulation systems. This interest has been motivated by many arguments, including the following: Natural language is a highly efficient means of interacting with non-linguistic AI systems. The other cognitive abilities, which are objects of interest of AI systems in general, are often directly or indirectly connected with linguistic abilities. Natural language systems are better suited to the expectations of naive users and can be managed by them better. Most natural-language artificial-intelligence systems have been developed as interfaces to some sort of task-oriented system. Typical of this function are question-answering systems involved with information storage and retrieval or other documentation tasks (WALTZ 1977).