Mining fuzzy-temporal gradual patterns

Abstract

Gradual patterns allow for retrieval of the correlations between attributes through rules such as “the more the exercise, the less the stress.” However, it may be the case that there is a lag between changes in some attributes and their impact on others. Current methods do not take this into account. In this paper, we extend existing methods to handle these situations in order to retrieve patterns such as: “the more the exercise increases, the more the stress decreases one month later”. We also extend our gradual rules to include fuzzy temporal constraints such as “the more the exercise increases, the more the stress decreases almost one month later”. For this kind of patterns, we designed three algorithms that were implemented and tested on real data.