Table 8 CoLog in comparison to other log anomaly detection methods on the Zookeeper dataset. Bold numbers indicate the outstanding results.
Anomaly detection technique | Precision | Recall | F1-Score | Accuracy |
|---|---|---|---|---|
Supervised Methods | ||||
Logistic Regression143 | 98.369 | 98.464 | 98.416 | 98.562 |
Support Vector Machines109 | 97.987 | 97.769 | 97.877 | 98.078 |
Decision Tree110 | 98.599 | 98.880 | 98.737 | 98.851 |
Attentional BiLSTM63 | 95.783 | 92.928 | 94.300 | 98.387 |
Convolutional Neural Network46 | 95.121 | 92.662 | 93.850 | 98.252 |
pylogsentiment38 | 99.722 | 99.898 | 99.810 | 99.973 |
Unsupervised Methods | ||||
Isolation Forest144 | 32.607 | 50.000 | 39.473 | 65.215 |
Principal Component Analysis12 | 79.011 | 51.209 | 42.121 | 66.021 |
LSTM36 | 95.825 | 91.887 | 93.750 | 98.252 |
Transformer80 | 99.890 | 99.416 | 99.652 | 99.361 |
CoLog | \(\textbf{100}\) | \(\textbf{100}\) | \(\textbf{100}\) | \(\textbf{100}\) |
