Posterior probability based ensemble strategy using optimizing decision directed acyclic graph for multi-class classification
作者： Ligang Zhou , Hamido Fujita
DOI: 10.1016/J.INS.2017.02.059
关键词: Heuristic 、 Data mining 、 Mathematics 、 Multiclass classification 、 Decomposition (computer science) 、 Ensemble learning 、 Decision rule 、 Binary number 、 Directed acyclic graph 、 Artificial intelligence 、 Machine learning 、 Posterior probability
摘要: Ensemble strategy is important to develop a decomposition and ensemble method for multi-class classification problems. Most existing strategies use predetermined heuristic decision rules. In this work, we build up the rules by optimizing directed acyclic graph (ODDAG) with classical fuzzy trees posterior probabilities of binary classifiers from one-vs-one (OVO) or one-vs-all (OVA) Four widely used extensible algorithms ten methods incorporating four (BCs) have been tested on 25 data sets. The empirical results show that based using ODDAG are among top achieve best performance in terms two different measures.
sci-hub.se 参考文章(40)
Sun-Phil Nah, Bo-Seok Seo, Woo-Hyun Ahn, Automatic Classification of Digitally Modulated Signals Based on K-Nearest Neighbor Springer, Dordrecht. pp. 63- 69 ,(2015) , 10.1007/978-94-017-9558-6_8
Bartosz Krawczyk, Michał Woźniak, Francisco Herrera, On the usefulness of one-class classifier ensembles for decomposition of multi-class problems Pattern Recognition. ,vol. 48, pp. 3969- 3982 ,(2015) , 10.1016/J.PATCOG.2015.06.001
Kevin P. Murphy, Machine Learning : A Probabilistic Perspective ,(2012)
Kevin Gurney, An introduction to neural networks ,(1997)
Thomas G. Dietterich, Ensemble Methods in Machine Learning Multiple Classifier Systems. pp. 1- 15 ,(2000) , 10.1007/3-540-45014-9_1
Janez Demšar, Statistical Comparisons of Classifiers over Multiple Data Sets Journal of Machine Learning Research. ,vol. 7, pp. 1- 30 ,(2006)
Koby Crammer, Yoram Singer, On the Learnability and Design of Output Codes for Multiclass Problems conference on learning theory. ,vol. 47, pp. 201- 233 ,(2002) , 10.1023/A:1013637720281
John C. Platt, Probabilistic Outputs for Support vector Machines and Comparisons to Regularized Likelihood Methods Advances in Large Margin Classifiers. ,(1999)
T. G. Dietterich, G. Bakiri, Solving multiclass learning problems via error-correcting output codes Journal of Artificial Intelligence Research. ,vol. 2, pp. 263- 286 ,(1994) , 10.1613/JAIR.105
Luís PF Garcia, José A Sáez, Julián Luengo, Ana C Lorena, André CPLF de Carvalho, Francisco Herrera, None, Using the One-vs-One decomposition to improve the performance of class noise filters via an aggregation strategy in multi-class classification problems Knowledge-Based Systems. ,vol. 90, pp. 153- 164 ,(2015) , 10.1016/J.KNOSYS.2015.09.023