Analyzing a psychiatric domain using a new knowledge discovery methodology

A new methodology for analyzing repeated very short-serial measurement on time in a medical ill-structured domain is introduced. This methodology is based on a combination of clustering based on rules with some Inductive Learning (AI) and clustering (Statistics) techniques. This proposal focuses on results obtained on a real application of this kind of data, where common statistical analysis (time series analysis, multivariate. . . ) and artificial intelligence techniques (knowledge based methods, inductive learning) of such data are often inadequate because of the intrinsic characteristics of those domains.