A novel bacterial foraging optimization algorithm for feature selection

作者：

Highlights：

• ACBFO and ISEDBFO are proposed based on original bacterial foraging optimization.

• The modified chemotaxis step raises selected probability of primary features in ACBFO.

• Swarming equation and elimination dispersal step are improved in ISEDBFO.

• ACBFO and ISEDBFO promote the classification accuracy and convergence speed.

• The proposed algorithms significantly outperformed other six metaheuristic algorithms.

摘要

•ACBFO and ISEDBFO are proposed based on original bacterial foraging optimization.•The modified chemotaxis step raises selected probability of primary features in ACBFO.•Swarming equation and elimination dispersal step are improved in ISEDBFO.•ACBFO and ISEDBFO promote the classification accuracy and convergence speed.•The proposed algorithms significantly outperformed other six metaheuristic algorithms.

论文关键词：Bacterial foraging optimization algorithm,Feature selection,Classification,Support vector machine

论文评审过程：Received 2 April 2016, Revised 6 April 2017, Accepted 7 April 2017, Available online 8 April 2017, Version of Record 20 April 2017.

论文官网地址：https://doi.org/10.1016/j.eswa.2017.04.019