Developing and Empirically Evaluating Robust Explanation Generators: The KNIGHT Experiments (1997)
James C. Lester, Bruce Porter

To explain complex phenomena, an explanation system must be able to select information from a formal representation of domain knowledge, organize the selected information into multi-sentential discourse plans, and realize the discourse plans in text. Although recent years have witnessed significant progress in the development of sophisticated computational mechanisms for explanation, empirical results have been limited. This paper reports on a seven year effort to empirically study explanation generation from semantically rich, large-scale knowledge bases. In particular, it describes KNIGHT, a robust explanation system that constructs multi-sentential and multi-paragrpah explanations from the Biology Knowledge Base, a large-scale knowledge base in the domain of botanical anatomy, physiology, and development. We introduce the Two Panel evaluation methodology and describe how KNIGHT's performance was assessed with this methodology in the most extensive empirical evaluation conducted of an explanation system. In this evaluation, KNIGHT scored within "half a grade" of domain experts, and its performance exceeded that of one of the domain experts.

Computational Linguistics Journal, Vol. 23, 1 (1997), pp. 65--101.

Peter Clark Formerly affiliated Research Scientist peterc [at] vulcan com
Bruce Porter Faculty porter [at] cs utexas edu