Research Article
Correlation-Based Anomaly Detection Method for Multi-sensor System
Table 6
Anomaly detection accuracy in terms of F1 score, on three datasets.
| Method | Dataset 1 | Dataset 2 | Dataset 3 | Mean | Standard deviation | Mean | Standard deviation | Mean | Standard deviation |
| Logistic regression | 0.87 | 0.007 | 0.82 | 0.035 | 0.94 | 0.006 | Decision tree | 0.91 | 0.006 | 0.83 | 0.086 | 0.97 | 0.011 | Naïve Bayes | 0.89 | 0.008 | 0.70 | 0.042 | 0.86 | 0.017 | QDA | 0.88 | 0.027 | 0.86 | 0.066 | 0.87 | 0.021 | SVM | 0.52 | 0.008 | 0.40 | 0.047 | 0.82 | 0.032 | CNN | 0.90 | 0.005 | 0.86 | 0.027 | 0.95 | 0.005 | Proposed method | 0.92 | 0.006 | 0.90 | 0.031 | 0.95 | 0.004 |
|
|
Bold font is to highlight the results of the proposed method.
|