English

Why only Micro-F1? Class Weighting of Measures for Relation Classification

Computation and Language 2022-05-20 v1 Machine Learning

Abstract

Relation classification models are conventionally evaluated using only a single measure, e.g., micro-F1, macro-F1 or AUC. In this work, we analyze weighting schemes, such as micro and macro, for imbalanced datasets. We introduce a framework for weighting schemes, where existing schemes are extremes, and two new intermediate schemes. We show that reporting results of different weighting schemes better highlights strengths and weaknesses of a model.

Keywords

Cite

@article{arxiv.2205.09460,
  title  = {Why only Micro-F1? Class Weighting of Measures for Relation Classification},
  author = {David Harbecke and Yuxuan Chen and Leonhard Hennig and Christoph Alt},
  journal= {arXiv preprint arXiv:2205.09460},
  year   = {2022}
}

Comments

NLP Power! The First Workshop on Efficient Benchmarking in NLP (ACL 2022)

R2 v1 2026-06-24T11:22:07.669Z