Recurring concept meta-learning for evolving data streams

作者：

Highlights：

• A fast, accurate classification framework for data streams that reuses classifiers.

• More accurate than a state-of-the-art classifier reuse framework on synthetic data.

• Five times faster and as accurate as this framework on real-world benchmarks.

• More accurate than a state-of-the-art ensemble technique on synthetic datasets.

• Six times faster and as accurate as this framework on real-world benchmarks.

摘要

•A fast, accurate classification framework for data streams that reuses classifiers.•More accurate than a state-of-the-art classifier reuse framework on synthetic data.•Five times faster and as accurate as this framework on real-world benchmarks.•More accurate than a state-of-the-art ensemble technique on synthetic datasets.•Six times faster and as accurate as this framework on real-world benchmarks.

论文关键词：Data streams,Concept drift,Recurring concepts,Classification

论文评审过程：Received 24 April 2019, Revised 19 July 2019, Accepted 20 July 2019, Available online 20 July 2019, Version of Record 25 July 2019.

论文官网地址：https://doi.org/10.1016/j.eswa.2019.112832