Skip to main content
. 2023 Jun 2;13:8989. doi: 10.1038/s41598-023-36172-7

Table 7.

Results for Power Users (downsampled), including standard deviation.

Power Users down
Accuracy F1-score AUC Precision Sensitivity Specificity
DT 0.734 (+/− 0.011) 0.738 (+/− 0.009) 0.739 (+/− 0.011) 0.729 (+/− 0.018) 0.748 (+/− 0.018) 0.720 (+/− 0.030)
RFC 0.773 (+/− 0.013) 0.773 (+/− 0.013) 0.855 (+/− 0.011) 0.775 (+/− 0.017) 0.771 (+/− 0.021) 0.775 (+/− 0.023)
SVM 0.684 (+/− 0.018) 0.685 (+/− 0.022) 0.722 (+/− 0.017) 0.682 (+/− 0.018) 0.689 (+/− 0.035) 0.679 (+/− 0.026)
CNB 0.563 (+/− 0.018) 0.536 (+/− 0.019) 0.622 (+/− 0.021) 0.572 (+/− 0.023) 0.504 (+/− 0.023) 0.621 (+/− 0.033)
KNC 0.760 (+/− 0.014) 0.757 (+/− 0.017) 0.827 (+/− 0.014) 0.766 (+/− 0.013) 0.749 (+/− 0.030) 0.771 (+/− 0.020)
LRC 0.608 (+/− 0.016) 0.566 (+/− 0.018) 0.665 (+/− 0.015) 0.635 (+/− 0.022) 0.511 (+/− 0.022) 0.706 (+/− 0.028)
MLP 0.730 (+/− 0.011) 0.725 (+/− 0.016) 0.797 (+/− 0.015) 0.739 (+/− 0.023) 0.715 (+/− 0.042) 0.745 (+/− 0.042)
XGB 0.777 (+/− 0.018) 0.772 (+/− 0.021) 0.857 (+/− 0.013) 0.788 (+/− 0.015) 0.757 (+/− 0.034) 0.796 (+/− 0.017)

Decision Tree (DT), Random Forest (RFC), Support Vector Machine (SVM), Complement Naive Bayes (CNB), k-nearest neighbors (KNC), Logistic Regression (LRC), Multi-layer Perceptron (MLP), Extreme Gradient Boosting (XGB)

Highest values are in bold.