作者: Jaime G Carbonell
DOI:
关键词:
摘要: Abstract : Interpreting metaphors is an integral and inescapable process in human understanding of natural language. Part I this paper discusses a method analyzing based on the existence small number generalized metaphor mappings. Each contains recognition network, basic mapping, additional transfer mappings, implicit intention component. It argued that reduces interpretation from reconstruction to task. Steps towards automating certain aspects language learning are also discussed. II analyzes analogical mappings underlying implications for inference memory organization. Regularities have been observed indicating types conceptual relations much more apt remain invariant than other relations, resulting induced invariance hierarchy. The central thesis processes governed by same manifest as (Author)