General Models of Usability 3 EFFICIENCY MEASURES MEASURES MINIMIZE COSTS QUALITATIVE SUCCESS MAXIMIZE TASK MAXIMIZE USER SATISFACTION

The design of methods for performance evaluation is a major open research issue in the area of spoken language dialogue systems. This paper presents the PARADISE methodology for developing predictive models of spoken dialogue performance, and shows how to evaluate the predictive power and generalizability of such models. To illustrate the methodology, we develop a number of models for predicting system usability (as measured by user satisfaction), based on the application of PARADISE to experimental data from three diierent spoken dialogue systems. We then measure the extent to which the models generalize across diierent systems, diierent experimental conditions, and diierent user populations, by testing models trained on a subset of the corpus against a test set of dialogues. The results show that the models generalize well across the three systems, and are thus a rst approximation towards a general performance model of system usability.