A hybrid model of fuzzy ARTMAP and genetic algorithm for data classification and rule extraction

作者：

Highlights：

• A hybrid model (QFAM-GA) for data classification and rule extraction is proposed.

• Fuzzy ARTMAP (FAM) with Q-learning is first used for incremental learning of data.

• A Genetic Algorithm (GA) is then used for feature selection and rule extraction.

• Pruning is used to reduce the network complexity and to facilitate rule extraction.

• The results show QFAM-GA can provide useful if-then rule to explain its predictions.

摘要

•A hybrid model (QFAM-GA) for data classification and rule extraction is proposed.•Fuzzy ARTMAP (FAM) with Q-learning is first used for incremental learning of data.•A Genetic Algorithm (GA) is then used for feature selection and rule extraction.•Pruning is used to reduce the network complexity and to facilitate rule extraction.•The results show QFAM-GA can provide useful if-then rule to explain its predictions.

论文关键词：Fuzzy ARTMAP,Reinforcement learning,Q-learning,Data classification,Rule extraction,Genetic algorithm

论文评审过程：Received 28 June 2015, Revised 11 September 2015, Accepted 13 November 2015, Available online 1 December 2015, Version of Record 5 January 2016.

论文官网地址：https://doi.org/10.1016/j.eswa.2015.11.009