A feature construction approach for genetic iterative rule learning algorithm

作者：

Highlights：

• A method to include feature construction in a fuzzy rule learning algorithm is proposed.

• The feature construction incorporates relations and functions in the antecedent of fuzzy rules.

• This procedure allows to increase the amount of information extracted from the initial variables.

• The proposal NSLV-FR obtains results with a good balance among accuracy, interpretability and time needed to get the model.

摘要

•A method to include feature construction in a fuzzy rule learning algorithm is proposed.•The feature construction incorporates relations and functions in the antecedent of fuzzy rules.•This procedure allows to increase the amount of information extracted from the initial variables.•The proposal NSLV-FR obtains results with a good balance among accuracy, interpretability and time needed to get the model.

论文关键词：Genetic fuzzy systems,Feature construction,Iterative learning approach,Classification

论文评审过程：Received 26 July 2012, Revised 3 December 2012, Accepted 14 March 2013, Available online 23 March 2013.

论文官网地址：https://doi.org/10.1016/j.jcss.2013.03.011