Intelligent interpretation of validation data

摘要

This paper describes the characteristics of an expert system that will make an intelligent interpretation of the quantitative data obtained following the usual validation process for intelligent systems. This expert system converts numerical data into information that will indicate whether the performance of the validated intelligent system is comparable to the performance of human experts from the domain taken as reference. The process is carried out in two parts: an algorithmic part which identifies the basic characteristics of the quantitative data; a heuristic part which analyses these characteristics so as to be able to draw conclusions as to the results of the system.