A general semantic model of negation in natural language: representation and inference
Lucja Iwańska · Principles of Knowledge Representation and Reasoning · 1992
The research contribution of this dissertation is a formal, computational model of the semantics of negation in natural language; its major components are two knowledge representation formalisms with Boolean semantics: a computable Boolean algebra for representing concepts and a computable Boolean algebra for representing the semantics of scalar predicate expressions of natural language that denote quantitative values. The UNO computer implementation demonstrates that this semantic model faithfully captures the logical inferences that people draw from English sentences with explicit negative information. The UNO natural language processing system is shown to correctly handle a substantial subset of English: knowledge from sentences involving negation as well as conjunction and disjunction of complex determiners, adjectives, adverbs, common nouns, proper nouns, noun phrases, verbs, verb phrases, prepositions, and prepositional phrases can be represented and reasoned with. Areas of future research involve extending the model to handle modal operators, sentential adverbs, pragmatics, temporal reasoning, intensionality and non-logical reasoning.