A Framework of Cognition and Conceptual Structures Based on Deep Semantics

Abstract

The argument is made that artificial intelligence (AI) systems which simulate cognition fail to reach significant goals because they rely on inadequate frameworks of cognition and conceptual structures that are based on shallow semantics. Shallow semantics examines only the surface meanings of expressions and words while deep semantics incorporates the meanings of the phonetic substructures of words. Shallow semantics is like chemistry that studies molecules (such as sugar) but never looks at elements (carbon, hydrogen, and oxygen). This paper briefly reviews three decades of deep semantics research by the author which produced a new framework of cognition and a new kind of conceptual structures. These have been validated by large-scale smart text retrieval and by constructing a theory of emotions and a consumer choice theory. Deep semantics is used to develop an ontology of cognition itself which is then contrasted with the cognitive frameworks and conceptual structures of AI systems.