This CRI planning grant uses the theory of conceptual metaphor to plan the creation of an on-line metaphor resource that covers multiple types of language and corpora. Metaphor is pervasive in human language and thought and is crucial for both linguistic expression and human reasoning. Conventionalized as well as novel metaphorical expressions can be routinely encountered in everyday life conversations, in newspapers, and in many other text types including scientific, technical, and literary writing.
The goal of this planning effort is to resolve issues related to corpus selection and preparation, and to incorporate community wide evaluative feedback in the iterative design and implementation of metaphor annotations. Towards the end of the planning work, there will be a workshop to evaluate the approach, to build support for the corpus creation enterprise, and to establish a user community. This corpus selection and metaphor annotation effort is also being coordinated with the companion SGER effort underway (NSF IIS grant 0843275) that is exploring issues related to scalable metaphor modeling and inference.
The multi-layer annotations will include conceptual relations and metaphor domains, linguistic expressions and frame annotations of the relevant lexical targets, and syntactic relations that obtain in the annotated target text. The result will be the design of a comprehensive, richly annotated resource of metaphoric languge.The combination of this work, along with parallel efforts in semantic analysis and inference should permit the development and testing of large-scale systems capable of figurative language processing within a few years.
