Skip to main content

Table 3 Balanced accuracy, Precision, and Recall, and F1 score for both malicious (Mal) and benign (Ben) accounts for supervised learning algorithms used in state of the art approaches where hyperparameters were reported

From: Detecting malicious accounts in permissionless blockchains using temporal graph properties

Reference

Classifier

Accuracy balanced

Precision

Recall

F1 score

MCC

Mal

Ben

Mal

Ben

Mal

Ben

Score

Ostapowicz and Zbikowski (2019)

RandomForest

0.64

0.98

1.00

0.29

1.00

0.44

1.00

0.52

SVM*

–

–

–

–

–

–

–

–

XGBoost

0.85

0.97

1.00

0.7

1.00

0.91

1.00

0.81

Singh (2019)

KNN

0.73

0.89

0.99

0.48

0.99

0.62

0.99

0.64

  1. Here, we use 59 features and both EOA and SC dataset configuration. Here, we also report the Matthews Correlation Coefficient (MCC) score
  2. *Did not converge as the dataset was large