On removing conflicts for machine learning
作者:
Highlights:
• Learning conflicts that affect machine learning system performance.
• Samples in a dataset with similar values but different target.
• Retention error kept meaningful samples while removing learning conflicts.
• Artificial neural networks trained with learning conflicts datasets.
• Generic algorithm improves the machine learning system performance.
摘要
•Learning conflicts that affect machine learning system performance.•Samples in a dataset with similar values but different target.•Retention error kept meaningful samples while removing learning conflicts.•Artificial neural networks trained with learning conflicts datasets.•Generic algorithm improves the machine learning system performance.
论文关键词:Learning conflicts,Training issues,Machine learning,Artificial neural networks,Genetic algorithm
论文评审过程:Received 15 February 2022, Revised 12 May 2022, Accepted 8 June 2022, Available online 15 June 2022, Version of Record 21 June 2022.
论文官网地址:https://doi.org/10.1016/j.eswa.2022.117835