COGPARSE: Brain-Inspired Knowledge-Driven Full Semantics Parsing - Radical Construction Grammar, Categories, Knowledge-Based Parsing & Representation

Humans use semantics during parsing; so should computers. In contrast to phrase structure-based parsers, COGPARSE seeks to determine which meaning-bearing components are present in a text, using world knowledge and lexical semantics for construction grammar form selection, syntactic overlap processing, disambiguation, and confidence calculation. In a brain-inspired way, COGPARSE aligns parsing with the structure of the lexicon, providing a linguistic representation, parsing algorithm, associated linguistic theory, and preliminary metrics for evaluating parse quality. Given sufficient information on nuanced word and construction semantics, COGPARSE can also assemble detailed full-semantics meaning representations of input texts. Beyond the ability to determine which parses are most likely to be intended and to use knowledge in disambiguation, full-semantics parsing enables nuanced meaning representation, learning, summarization, natural language user interfaces, and the taking of action based on natural language input. 1 Towards a Next Generation of Brain-Inspired Parsers COGPARSE is a natural language parsing framework combining semantics and world knowledge with syntax, demonstrating brain-inspired natural language processing (NLP) along the way. Its ultimate goal is to enable full-semantics processing, defined here as both 1) the generation of highly nuanced semantic representations from input texts and 2) the use of semantics to directly inform parsing. COGPARSE offers a unique way of thinking about what parsing is and how it should be done. Instead of attempting to determine ‘correct parses’ grounded in phrase structures, COGPARSE discovers the meaning-bearing components present in a text, using lexical semantics and world knowledge for linguistic form selection, overlap detection, and disambiguation. Viewing texts in terms of meaning-bearing components enables NLP systems to much more closely and easily align text spans with the meaning they contribute to the overall whole. It allows parsers to process text more like humans do (as modeled by cognitive linguistics), and to create systems that process inputs at the same level of syntactic and semantic granularity at which they were H. Zhang et al. (Eds.): BICS 2012, LNAI 7366, pp. 1–11, 2012. c © Springer-Verlag Berlin Heidelberg 2012