Analysis of diversity methods for evolutionary multi-objective ensemble classifiers
Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt
Ensemble classifiers are strong and robust methods for classification and regression tasks. Considering the balance between runtime and classifier accuracy the learning problem becomes a multi-objective optimization problem. In this work, we propose an evolutionary multiobjective algorithm based on non-dominated sorting that balances runtime and accuracy properties of nearest neighbor classifier ensembles and decision tree ensembles. We identify relevant ensemble parameters with a significant impact on the accuracy and runtime. In the experimental part of this paper, we analyze the behavior on typical classification benchmark problems.
Originalsprog | Engelsk |
---|---|
Titel | Applications of Evolutionary Computation - 18th European Conference, EvoApplications 2015, Proceedings |
Redaktører | Giovanni Squillero, Antonio M. Mora |
Antal sider | 12 |
Forlag | Springer Verlag, |
Publikationsdato | 1 jan. 2015 |
Sider | 567-578 |
ISBN (Elektronisk) | 9783319165486 |
DOI | |
Status | Udgivet - 1 jan. 2015 |
Eksternt udgivet | Ja |
Begivenhed | 18th European Conference on the Applications of Evolutionary Computation, EvoApplications 2015 - Copenhagen, Danmark Varighed: 8 apr. 2015 → 10 apr. 2015 |
Konference
Konference | 18th European Conference on the Applications of Evolutionary Computation, EvoApplications 2015 |
---|---|
Land | Danmark |
By | Copenhagen |
Periode | 08/04/2015 → 10/04/2015 |
Sponsor | Institute for Informatics and Digital Innovation, National Museum of Denmark, The World Federation on Soft Computing |
Navn | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
---|---|
Vol/bind | 9028 |
ISSN | 0302-9743 |
ID: 223196683