Most rule-interest measures are suitable for binary attributes and using an unsupervised usual algorithm for the discovery of association rules requires a transformation for other kinds of attributes. Given that the complexity of these algorithms increases exponentially with the number of attributes, this transformation can lead us, on the one hand to a combinatorial explosion, and on the other hand to a prohibitive number of weakly significant rules with many redundancies. To fill the gap, we propose in this study a new objective rule-interest measure called intensity of inclination which evaluates the implication between two ordinal attributes (numeric or ordinal categorical attributes). This measure allows us to extract a new kind of knowledge: ordinal association rules. An evaluation of an application to some banking data ends up the study.