Towards Modeling Data Quality and Machine Learning Model Performance

2024/12/08 by Usman Anjum, Anjum, Usman, Chris Trentman +4
Computer Science · Decision Sciences · #Advanced Database Systems and Queries #Artificial Intelligence (cs.AI) #Data Mining Algorithms and Applications #Data Quality and Management #FOS: Computer and information sciences #Machine Learning (cs.LG)

paper · pdf · doi:10.48550/arxiv.2412.05882

openalex publication_date 2024/12/08 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Understanding the effect of uncertainty and noise in data on machine learning models (MLM) is crucial in developing trust and measuring performance. In this paper, a new model is proposed to quantify uncertainties and noise in data on MLMs. Using the concept of signal-to-noise ratio (SNR), a new metric called deterministic-non-deterministic ratio (DDR) is proposed to formulate performance of a model. Using synthetic data in experiments, we show how accuracy can change with DDR and how we can use DDR-accuracy curves to determine performance of a model.

Related

• Quantifying Uncertainty and Variability in Machine Learning: Confidence Intervals for Quantiles in Performance Metric Distributions
• Confidence-based Estimators for Predictive Performance in Model Monitoring
• Towards Inferential Reproducibility of Machine Learning Research
• Quality of Data in Machine Learning
• Learning Prediction Intervals for Model Performance