# Novel split quality measures for stratified multilabel Cross Validation with application to large and sparse gene ontology datasets

arXiv: https://arxiv.org/abs/2109.01425 · HTML: https://mlsanity.com/paper/2109.01425/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 5 of its 15 references that are in the map.
- First posted on arXiv 3 Sep 2021.

## Builds on (most cited first, 5 of 5)

- [Mean Absolute Percentage Error for regression models](https://mlsanity.com/paper/1605.02541/) (2016)
- [A Network Perspective on Stratification of Multi-Label Data](https://mlsanity.com/paper/1704.08756/) (2017)
- [A scikit-based Python environment for performing multi-label classification](https://mlsanity.com/paper/1702.01460/) (2017)
- [Stratified Sampling for Extreme Multi-Label Data](https://mlsanity.com/paper/2103.03494/) (2021)
- [EvoSplit: An evolutionary approach to split a multi-label data set into disjoint subsets](https://mlsanity.com/paper/2102.06154/) (2021)

## Built on by (most cited first, 3 of 3)

- [Stratify or Die: Rethinking Data Splits in Image Segmentation](https://mlsanity.com/paper/2509.21056/) (2025)
- [Cardi-GPT: An Expert ECG-Record Processing Chatbot](https://mlsanity.com/paper/2510.24737/) (2025)
- [Not All Nodes Are Created Equal: Homophily-Aware Stratification for Stable GNN Evaluation](https://mlsanity.com/paper/2609.19210/) (2026)

## Abstract (arXiv)

Multilabel learning is an important topic in machine learning research. Evaluating models in multilabel settings requires specific cross validation methods designed for multilabel data. In this article, we show that the most widely used cross validation split quality measures do not behave adequately with multilabel data that has strong class imbalance. We present improved measures and an algorithm, optisplit, for optimising cross validations splits. We present an extensive comparison of various types of cross validation methods in which we show that optisplit produces more even cross validation splits than the existing methods and that it is fast enough to be used on big Gene Ontology (GO) datasets.