Sample Size Requirements for Popular Classification Algorithms in Tabular Clinical Data: Empirical Study
Scott Silvey, Jinze Liu · 2024
Read Sample Size Requirements for Popular Classification Algorithms in Tabular Clinical Data: Empirical Study by Scott Silvey, Jinze Liu (2024) 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-11688588.txt.
Description
Scott Silvey and Jinze Liu ask how much data clinical classifiers really need in a 2024 study. Every prediction model improves with more cases, then flattens out at some point. Past that point, extra data costs money and time without better answers.
Sixteen large open clinical datasets with yes-or-no outcomes fed four common algorithms. The set covered boosted trees, random forests, logistic regression, and neural networks. Cross-validated accuracy was tracked as sample size grew for each pairing. Dataset traits that predict the flattening point were also sought. Study planners can use such curves to size a trial once, well, instead of guessing or overspending.
Catalog record
| CREATOR | Scott Silvey, Jinze Liu |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 40 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |