Research on ordinal regression of interval valued data

Traditional regression algorithms can only deal with non-ordinal regression problems of crisp data. In order to deal with the problem of the regression of interval valued data with the monotonic constraint, an algorithm is proposed in this paper, where the condition attributes are ordinal interval values and the decision attribute is continuous. The proposed algorithm employs the combination of the variance and rank mutual information as heuristic. The advantage of the proposed algorithm is that it can consider both the dispersion degree of the decision attribute value and the monotonie consistency degree that between condition attributes and decision attribute.