Dynamic selection of classifiers based on complexity measures
Ederson Schmeing, André Luiz Brun, Ronan Assumpção Silva · 2022
Multiple Classifier Systems (MCS) have been an alternative to the monolithic approach to improve the performance of the classification task and increase the accuracy of pattern recognition. The dynamic selection (DS) has shown to be a promising strategy. The architecture in which classification systems are built influences their performance. However, the difficulty of the classification problem is usually neglected. Studies show that a better understanding of data complexity can be interesting for classifier selection. Thus, in this work, we evaluate the complexity measures as selection criteria for the dynamic selection of classifiers. Modifying classical literature approaches results in higher accuracy of 39.51% of the cases by adding the complexity measures while maintaining the original approach corresponds to 28.4%. Ties were present in 35.8%. This work also presents the information gain of 14 complexity measures, which suggest some drawbacks of using some of them.