Designing multi-label classifiers that maximize F measures: State of the art

作者：

Highlights：

• We survey classification algorithms that maximize F measures.

• We consider the empirical utility maximization and decision-theoretic approaches.

• We consider first the single-label F measure.

• We then consider the multi-label instance-wise, macro- and micro-averaged F.

摘要

Highlights•We survey classification algorithms that maximize F measures.•We consider the empirical utility maximization and decision-theoretic approaches.•We consider first the single-label F measure.•We then consider the multi-label instance-wise, macro- and micro-averaged F.

论文关键词：Multi-label classification,F measure,Learning algorithms,Empirical utility maximization,Decision-theoretic approach

论文评审过程：Received 22 February 2016, Revised 15 July 2016, Accepted 10 August 2016, Available online 13 August 2016, Version of Record 26 August 2016.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.08.008