| Title |
Sensitivity versus accuracy in multiclass problems using memetic pareto evolutionary neural networks |
| Authors |
Fernández Caballero J.C. , MARTÍNEZ ESTUDILLO, FRANCISCO JOSÉ, Hervas C. , Gutierrez P.A. |
| External publication |
No |
| Means |
IEEE TRANSACTIONS ON NEURAL NETWORKS |
| Scope |
Article |
| Nature |
Científica |
| JCR Quartile |
1 |
| JCR Impact |
2.633 |
| Web |
https://www.scopus.com/inward/record.uri?eid=2-s2.0-77951937354&doi=10.1109%2fTNN.2010.2041468&partnerID=40&md5=1d9c4d0a943ad67464d913552e4804db |
| Publication date |
01/01/2010 |
| ISI |
000277337200004 |
| Scopus Id |
2-s2.0-77951937354 |
| DOI |
10.1109/TNN.2010.2041468 |
| Abstract |
This paper proposes a multiclassification algorithm using multilayer perceptron neural network models. It tries to boost two conflicting main objectives of multiclassifiers: a high correct classification rate level and a high classification rate for each class. This last objective is not usually optimized in classification, but is considered here given the need to obtain high precision in each class in real problems. To solve this machine learning problem, we use a Pareto-based multiobjective optimization methodology based on a memetic evolutionary algorithm. We consider a memetic Pareto evolutionary approach based on the NSGA2 evolutionary algorithm (MPENSGA2). Once the Pareto front is built, two strategies or automatic individual selection are used: the best model in accuracy and the best model in sensitivity (extremes in the Pareto front). These methodologies are applied to solve 17 classification benchmark problems obtained from the University of California at Irvine (UCI) repository and one complex real classification problem. The models obtained show high accuracy and a high classification rate for each class. © 2010 IEEE. |
| Keywords |
Bench-mark problems; Best model; Classification ,; Classification rates; Evolutionary approach; Evolutionary neural network; High precision; Local search; Machine learning problem; Memetic; Memetic evolutionary algorithms; Multi objective evolutionary algorithms; Multi-class problems; Multi-classification; Multi-layer perceptron neural networks; Multiclassifiers; Pareto front; Real problems; University of California; Learning algorithms; Multiobjective optimization; Neural networks; Pareto principle; Evolutionary algorithms; algorithm; article; artificial intelligence; artificial neural network; biological model; computer network; computer simulation; evolution; genetics; human; mutation; receiver operating characteristic; reproducibility; statistics; Algorithms; Artificial Intelligence; Computer Communication Networks; Computer Simulation; Evolution; Humans; Models, Genetic; Mutation; Neural Networks (Computer); Reproducibility of Results; ROC Curve |
| Universidad Loyola members |
|
Manage cookies