Critical Challenges and Guidelines in Evaluating Synthetic Tabular Data: A Systematic Review
Nazia Nafis, Iñaki Esnaola, Alvaro Martinez-Perez, Maria-Cruz Villa-Uriol, Venet Osmani · 2025
Read Critical Challenges and Guidelines in Evaluating Synthetic Tabular Data: A Systematic Review by Nazia Nafis, Iñaki Esnaola, Alvaro Martinez-Perez, Maria-Cruz Villa-Uriol, Venet Osmani (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/arxiv-2504-18544.txt.
Description
Nazia Nafis and co-authors wrote a 2025 systematic review on how to judge fake health data made by computer models. Synthetic tables can stand in for real patient records in research, but poor fakes can mislead. Judging their quality turns out to be as hard as making them.
The team screened 2,067 papers from the past ten years and kept 134 for close study. They found no shared standard for judging fake data, uneven use of scores, and little input from health experts. Papers also skipped key facts about their data and were hard to repeat. The review groups methods into clear families and gives practical rules for stronger checks. The aim is safer, more open use of synthetic health data.
Catalog record
| CREATOR | Nazia Nafis, Iñaki Esnaola, Alvaro Martinez-Perez, Maria-Cruz Villa-Uriol, Venet Osmani |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 59 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-18544