A fuzzy evolutionary framework for combining ensembles

详细信息    查看全文

• 作者：Athanasios Tsakonas ; ^{thanos.tsakonas@gmail.com} ; ^{atsakonas@bournemouth.ac.uk} ; Bogdan Gabrys ^{bgabrys@bournemouth.ac.uk}
• 关键词：Ensemble systems ; Function approximation ; Fuzzy rule based systems ; Genetic programming
• 刊名：Applied Soft Computing Journal
• 出版年：2013
• 出版时间：April, 2013
• 年：2013
• 卷：13
• 期：4
• 页码：1800-1812
• 全文大小：1909 K

文摘

We propose an evolutionary framework for the production of fuzzy rule bases where each rule executes an ensemble of predictors. The architecture, the rule base and the composition of the ensembles are evolved over time. To achieve this, we employ a context-free grammar within a hybrid genetic programming system using a multi-population model. As base predictors, multilayer perceptron neural networks and support vector machines are available. We apply the system to several function approximation and regression tasks and compare the results with recent research and state-of-the-art models. We conclude that the proposed architecture is competitive and has a number of very desirable features supporting automation of predictive model building and their adaptation over time. Finally, we suggest further potential research directions.