SODE: Self-Adaptive One-Dependence Estimators for classification
作者：
Highlights：
• Self-adaptive attribute weighting for One-Dependence Estimators.
• Artificial immune systems (AIS) for attribute weighting.
• Combining learning objective and AIS affinity function for attribute weighting.
• Experiments on 58 real-world datasets demonstrating performance gain.
• Trade-off between runtime efficiency and accuracy effectiveness.
摘要
Highlights•Self-adaptive attribute weighting for One-Dependence Estimators.•Artificial immune systems (AIS) for attribute weighting.•Combining learning objective and AIS affinity function for attribute weighting.•Experiments on 58 real-world datasets demonstrating performance gain.•Trade-off between runtime efficiency and accuracy effectiveness.
论文关键词：Attribute weighting,Self-adaptive,Naive Bayes,Classification,Artificial immune systems,Evolutionary machine learning
论文评审过程：Received 7 January 2015, Revised 6 August 2015, Accepted 25 August 2015, Available online 5 September 2015, Version of Record 27 November 2015.
论文官网地址：https://doi.org/10.1016/j.patcog.2015.08.023
