Balancing Exploration and Exploitation: A novel active learner for imbalanced data

作者：

Highlights：

• Active learning aims to reduce the amount of time and cost for labeling data.

• Active learner selects the smallest possible training data for learning a model.

• Our active learner aims to select the most informative and representative points.

• The proposed active learner obtained competitive results with the imbalanced data.

摘要

•Active learning aims to reduce the amount of time and cost for labeling data.•Active learner selects the smallest possible training data for learning a model.•Our active learner aims to select the most informative and representative points.•The proposed active learner obtained competitive results with the imbalanced data.

论文关键词：Active learning,Queries,Imbalanced data,Pool-based model

论文评审过程：Received 7 June 2020, Revised 10 August 2020, Accepted 30 September 2020, Available online 8 October 2020, Version of Record 9 October 2020.

论文官网地址：https://doi.org/10.1016/j.knosys.2020.106500