USING POLYNOMIAL APPROXIMATIONS TO DISCOVER QUALITATIVE MODELS
Reha Kamil Gerceker · 2006
Automating the discovery of qualitative models from observations is a difficult problem of machine learning and various algorithms have been proposed for the solution of this problem in the literature. In this paper, we present a new algorithm called LYQUID, which uses polynomials fitted on observed numerical data as approximations to the underlying real world functions; constraint discovery is then performed over those polynomials. LYQUID is shown to be a fast and successful learning algorithm even in the presence of a high level of noise.