Table 7 Model calibration: summary of lower and upper bound ranges for different diseases and datasets by year periods: 2010–2014, 2015–2019, 2020–2024.
From: Detection of ChatGPT fake science with the xFakeSci learning algorithm
Alzheimer’s | Cancer | Depression | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
GPT | Pub 10–14 | 15–19 | 20–24 | GPT | 10–14 | 15–19 | 20–24 | GPT | 10–14 | 15–19 | 20–24 | |
Lower | 0.27 | 0.15 | 0.14 | 0.14 | 0.25 | 0.13 | 0.15 | 0.15 | 0.28 | 0.10 | 0.09 | 0.10 |
Upper | 0.30 | 0.16 | 0.15 | 0.15 | 0.29 | 0.15 | 0.16 | 0.17 | 0.32 | 0.11 | 0.13 | 0.11 |
