Log Anomaly Detection on BlueGene/L
99.999PrecisionCoLog
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| CoLogLearning Strategy=Supervised2025.12 | 99.999 | 99.99 | 99.994 | 99.998 | |
| pylogsentimentLearning Strategy=Supervised2025.12 | 99.892 | 99.963 | 99.928 | 99.98 | |
| Attentional BiLSTMLearning Strategy=Supervised2025.12 | 97.64 | 97.955 | 97.792 | 97.902 | |
| Convolutional Neural NetworkLearning Strategy=Supervised2025.12 | 97.64 | 97.955 | 97.792 | 97.902 | |
| TransformerLearning Strategy=Unsupervised2025.12 | 97.64 | 97.955 | 97.792 | 97.902 | |
| LSTMLearning Strategy=Unsupervised2025.12 | 97.414 | 98.296 | 97.806 | 97.902 | |
| Decision TreeLearning Strategy=Supervised2025.12 | 60.576 | 50.998 | 50.303 | 92.348 | |
| Logistic RegressionLearning Strategy=Supervised2025.12 | 54.028 | 51.852 | 52.092 | 90.368 | |
| Isolation ForestLearning Strategy=Unsupervised2025.12 | 53.081 | 50.047 | 51.519 | 47.389 | |
| Principal Component AnalysisLearning Strategy=Unsupervised2025.12 | 51.168 | 54.26 | 38.97 | 48.487 | |
| Support Vector MachinesLearning Strategy=Supervised2025.12 | 46.314 | 50 | 48.087 | 92.628 |