Saved in:
Bibliographic Details
Main Authors: Anjum, Usman, Trentman, Chris, Caden, Elrod, Zhan, Justin
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2412.05882
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915053413859328
author Anjum, Usman
Trentman, Chris
Caden, Elrod
Zhan, Justin
author_facet Anjum, Usman
Trentman, Chris
Caden, Elrod
Zhan, Justin
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2412_05882
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Towards Modeling Data Quality and Machine Learning Model Performance
Anjum, Usman
Trentman, Chris
Caden, Elrod
Zhan, Justin
Machine Learning
Artificial Intelligence
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.
title Towards Modeling Data Quality and Machine Learning Model Performance
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2412.05882