An evolutionary approach for finding existingrelationships among several variables of a multidimensionaltime series is presented in this work. The proposed model todiscover these relationships is based on quantitative associationrules. This algorithm, called QARGA (QuantitativeAssociation Rules by Genetic Algorithm), uses a particularcodification of the individuals that allows solving two basicproblems. First, it does not perform a previous attributediscretization and, second, it is not necessary to set whichvariables belong to the antecedent or consequent. Therefore,it may discover all underlying dependencies amongdifferent variables. To evaluate the proposed algorithmthree experiments have been carried out. As initial step,several public datasets have been analyzed with the purposeof comparing with other existing evolutionary approaches.Also, the algorithm has been applied to synthetic time series(where the relationships are known) to analyze its potentialfor discovering rules in time series. Finally, a real-worldmultidimensional time series composed by several climatologicalvariables has been considered. All the results showa remarkable performance of QARGA.
展开▼
