Table 10 Comparing the results with earlier studies.
From: Using machine learning algorithms to enhance IoT system security
Accuracy (%) | Detection rate | FPR | Training time | |
|---|---|---|---|---|
Ambusaidi et al.38 | 92.3 | 92.3% | 0.41 | – |
Moustafa et al.39 | 97.8 | 97.8% | 2.5% | – |
Tsai et al.40 | 96.9 | 97.8% | 2.5% | – |
Alom et al.41 | 97.5 | – | – | 3.2 s |
Yin et al.42 | 99.5 | 97.1% | 3.6% | 5516 s |
Tang et al.43 | 75.8 | 75.0% | 15% | – |
Ludwig44 | 92.5 | 98.0% | 14.7% | – |
Al-Hawawreh et al.45 | 98.6 | 99.0% | 1.8% | 398 s |
Shone et al.46 | 98.0 | 71.0% | - | – |
Subba et al.47 | 97.6 | 97.3% | - | – |
The proposed app (experiment #1) | 99.1 | 97.3% | 0.2% | 425.5 |
The proposed app (experiment #2) | 99.5 | 98.6% | 0.7% | 230.4 |
The proposed app (experiment #3) | 99.9 | 99.8% | 0.1% | 192.7 |
