Upper Modeling: A general organization of knowledge for natural language processing

A general, reusable computational resource has been developed within the Penman text generation project for organizing domain knowledge appropriately for linguistic realization. This resource, called the upper model, provides a domainand task-independent classi cation system that supports sophisticated natural language processing while signi cantly simplifying the interface between domain-speci c knowledge and general linguistic resources. Although levels of abstract semantic organization similar to the upper model are now being sought in many natural language systems, often with a view to exploring appropriate formalization techniques, these mostly su er from either a lack of theoretical constraint concerning their internal contents and organization and the necessary mappings between them and surface realization, or a lack of abstraction which binds them too closely with linguistic form. This paper presents the results of our experiences in designing and using the upper model in a variety of applications over the past 5 years. In particular, we present our conclusions concerning the appropriate organization of an upper model, its domain-independence, and the types of interrelationships that need to be supported between upper model and grammar and semantics. We consider the organization and interrelationships that we have found necessary both useful for organizing domain knowledge consistently and vital for sophisticated generation capabilities. 1 Financial support was provided by AFOSR contract F49620-87-C-0005, and in part by DARPA contract MDA903-87-C-641. The opinions in this report are solely those of the author.