Skip to main content

Advertisement

Table 5 Quality of machine learning suggestions: Precision/recall of the machine learning text highlights with regard to the users’ final edits (calculated for each bias assessment which used machine learning assistance, and averaged)

From: Machine learning to help researchers evaluate biases in clinical trials: a prospective, randomized user study

  Overall (n = 328) RSG (n = 82) AC (n = 82) BPP (n = 82) BOA (n = 82)
Recall (SD) 0.90 (0.14) 0.91 (0.18) 0.92 (0.17) 0.90 (0.19) 0.91 (0.21)
Precision (SD) 0.87 (0.21) 0.91 (0.21) 0.88 (0.24) 0.87 (0.25) 0.87 (0.27)