CAREER: Automatic Knowledge Acquisition for Natural Language Interfaces to Educational Applications

University Of Illinois At Chicago, Chicago IL

Investigators

Abstract

This is a Faculty Early Career Development (CAREER) award. The project is designed to develop fundamental principles for natural language interfaces for the next generation of intelligent tutoring systems. Natural language interfaces to instructional and educational applications need both knowledge about how to act in a domain and communication knowledge about how to talk about acting in the domain. This project seeks to overcome the domain dependence of natural language interfaces that prevents them from being portable to other domains. It will accomplish this in two ways. First, the research will elucidate the difference between domain dependent and domain independent communication strategies. Second, it will develop a methodology, algorithms and tools to semi-automatically derive domain and communication knowledge from text and dialogues. An innovative aspect of this work is the role played by the lexical semantics of action verbs to acquire domain knowledge represented as action schemas. Among the instructional domains addressed by this research is computer science. The fact that the work concerns tutoring systems integrates the research and education components of the work, through class evaluation and undergraduate participation in research. This CAREER award recognizes and supports the early career-development activities of a teacher-scholar who is likely to become an academic leader of the twenty-first century. Success in this research will enable the construction of improved computerized tutors, as well as advancing the general field of natural language processing. A specific educational focus is mentoring women at the undergraduate and graduate level, and supporting girls in choosing science and engineering as a career.