Representing American Sign Language Classifier Predicates Using Spatially Parameterized Planning Templates

T his chapter addresses the concept of generalization primarily from a lin-guistic and engineering standpoint. We examine this issue from the perspective of the field of natural language processing, which explores how computer software can be created to automatically understand or generate information in the form of human languages. Our research program focuses on the creation of software that can generate animations of American Sign Language (ASL); specifically, we have examined how to best design software for translating from English sentences into animations of a virtual human character performing ASL. Our motivation for building this software is to make information more accessible to people who are deaf. A majority of deaf 18-year-olds in the United States have an English reading level below that of a typical 10-year-old student, and so machine translation software that could translate English text into ASL animations could significantly improve these individuals’ access to information, communication, and services.This chapter explores how the choice of representation for the linguistic structure of a specific construction in a language can have an impact on our ability to build software to generate that construction. Specifically, this chapter will focus on computational linguistic software for generating animations of a kind of ASL sentence called a classifier predicate (CP), a phenomenon in which signers use special hand movements to indicate the location and movement of invisible objects