# Optimal Ratio for Data Splitting

arXiv: https://arxiv.org/abs/2202.03326 · HTML: https://mlsanity.com/paper/2202.03326/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 28 references that are in the map.
- First posted on arXiv 7 Feb 2022.

## Builds on (most cited first, 4 of 4)

- [Support points](https://mlsanity.com/paper/1609.01811/) (2016)
- [SPlit: An Optimal Method for Data Splitting](https://mlsanity.com/paper/2012.10945/) (2020)
- [Data Twinning](https://mlsanity.com/paper/2110.02927/) (2021)
- [Test Set Sizing Via Random Matrix Theory](https://mlsanity.com/paper/2112.05977/) (2021)

## Built on by (most cited first, 18 of 18)

- [Data Readiness for AI: A 360-Degree Survey](https://mlsanity.com/paper/2404.05779/) (2024)
- [Model Reporting for Certifiable AI: A Proposal from Merging EU Regulation into AI Development](https://mlsanity.com/paper/2307.11525/) (2023)
- [The Impact of Feature Scaling In Machine Learning: Effects on Regression and Classification Tasks](https://mlsanity.com/paper/2506.08274/) (2025)
- [On Real-time Image Reconstruction with Neural Networks for MRI-guided Radiotherapy](https://mlsanity.com/paper/2202.05267/) (2022)
- [Survey and Taxonomy: The Role of Data-Centric AI in Transformer-Based Time Series Forecasting](https://mlsanity.com/paper/2407.19784/) (2024)
- [A Flexible Cell Classification for ML Projects in Jupyter Notebooks](https://mlsanity.com/paper/2403.07562/) (2024)
- [On the Cost of Model-Serving Frameworks: An Experimental Evaluation](https://mlsanity.com/paper/2411.10337/) (2024)
- [Revolutionizing Communication with Deep Learning and XAI for Enhanced Arabic Sign Language Recognition](https://mlsanity.com/paper/2501.08169/) (2025)
- [Category-based Galaxy Image Generation via Diffusion Models](https://mlsanity.com/paper/2506.16255/) (2025)
- [Exploring the Effectiveness of Dataset Synthesis: An application of Apple Detection in Orchards](https://mlsanity.com/paper/2306.11763/) (2023)
- [Deep Convolutional Neural Networks for Short-Term Multi-Energy Demand Prediction of Integrated Energy Systems](https://mlsanity.com/paper/2312.15497/) (2023)
- [Learning Run-time Safety Monitors for Machine Learning Components](https://mlsanity.com/paper/2406.16220/) (2024)
- [Integrating Boosted learning with Differential Evolution (DE) Optimizer: A Prediction of Groundwater Quality Risk Assessment in Odisha](https://mlsanity.com/paper/2502.17929/) (2025)
- [Test Set Sizing for the Ridge Regression](https://mlsanity.com/paper/2504.19231/) (2025)
- [Deep Learning in Renewable Energy Forecasting: A Cross-Dataset Evaluation of Temporal and Spatial Models](https://mlsanity.com/paper/2505.03109/) (2025)
- [Discovery of Fatigue Strength Models via Feature Engineering and automated eXplainable Machine Learning applied to the welded Transverse Stiffener](https://mlsanity.com/paper/2507.02005/) (2025)
- [Leveraging Generative AI for Enhancing Domain-Driven Software Design](https://mlsanity.com/paper/2601.20909/) (2026)
- [PRAISE: A Reproducible Workflow for Non-Programmer Scientists to Implement Published AI Models](https://mlsanity.com/paper/2610.16980/) (2026)

## Abstract (arXiv)

It is common to split a dataset into training and testing sets before fitting a statistical or machine learning model. However, there is no clear guidance on how much data should be used for training and testing. In this article we show that the optimal splitting ratio is √p:1, where p is the number of parameters in a linear regression model that explains the data well.