No abstract

This work presents a novel approach to multivariate time series classification. The method exploits the multivariate structure of the time series and the possibilities of the stacking ensemble method. The basics of the method may be described in three steps: first, decomposing the multivariate time series on its constituent univariate time series; second, inducing a classifier for each univariate time series plus and additional multivariate classifier for the whole time series; third, creating the final multivariate time series classifier stacking the previous classifiers. The ensemble obtained has the potential to improve the accuracy of the single multivariate time series classifier. Several configurations of the stacking method have been tested on seven multivariate time series data sets. In five out of seven data sets, the proposed method obtains the smallest error rate. Moreover, in two out of seven data sets, stacking only the univariate time series classifiers provides the best results. The experimental results show that when a multivariate time series method does not produce an accurate classifier, stacking it with univariate time series classifiers is an alternative worthy of consideration.

No abstract

Abstract. Diagnosis of real world problems demands the integration of different techniques from several research fields. In Model-based Diagnosis, both Artificial Intelligence and Control Theory communities have provided different but complementary approaches. Recent works, known as BRIDGE proposal, provided a common framework for the integration of techniques for static systems.This work proposes the extension of the BRIDGE framework for a specific class of dynamic systems, thus analyzing the influence of dynamic constraints in the behavior estimation capabilities for two Model-based Diagnosis techniques: Possible Conflicts and Analytical Redundancy Relations obtained through structural analysis. Results show the strong similarities between them, and provide new ways for integration of techniques from both areas. Additionally, algorithms computing Possible Conflicts provide the implicit structural model for state observer design with no extra knowledge added in the model. Results on a case study are provided, then compared and discussed against existing proposals.

No abstract