The paper presents an algorithm for finding the value of a discrete attribute which optimal splits a set of examples into two subsets, where the optimality is defined with regard to a criterion which is not computed over individual examples (like entropy) but on pairs of examples. The algorithm reduces the time complexity from quadratic (which we would get with a brute force algorithm) to linear. This is important for qualitative modelling where qualitative change vectors (as used in QUIN) are defined on pairs of examples and not on single examples.
COBISS.SI-ID: 7550548
The paper describes the methods of qualitative induction developed in early stages of the project: the first triangle method, which uses interpolation within a single simplex of Delaunay triangulation, star regression method, which computes linear regression within a star (a set of simplices around the reference point) and the tube regression method which computes linear regression in a hyper-tube along the axis of differentiation. The paper describes methods which we subsequently improved and replaced with better methods mostly based on the tube regression.
COBISS.SI-ID: 7625812