Assessing the Impact of the Quality of Textual Data on Feature Representation and Machine Learning Models: Quantitative Study Using Large Language Models

Tabinda Sarwar, Antonio José Jimeno Yepes, Lawrence Cavedon · 2025

Read Assessing the Impact of the Quality of Textual Data on Feature Representation and Machine Learning Models: Quantitative Study Using Large Language Models by Tabinda Sarwar, Antonio José Jimeno Yepes, Lawrence Cavedon (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-12811037.txt.

Description

Tabinda Sarwar and co-authors measured how text quality bends health AI in a 2025 study. Clinic notes hold rich facts about patients, but real notes come with typos, gaps, and shorthand, unlike clean trial data. A model that learns from shabby text can mispredict, and in health care a wrong call can cost a life.

The team varied the quality of note text in set steps and watched both the learned features and the model scores shift. Large language models helped read the notes, which made the test current. Their measures show how much dirt a model can bear. The work guides data care before models ship.

Catalog record

CREATORTabinda Sarwar, Antonio José Jimeno Yepes, Lawrence Cavedon
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE62 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC12811037

Browse the full library · About the Tower of Babel