Research Article
A Buffer Overflow Prediction Approach Based on Software Metrics and Machine Learning
Table 6
Decision tree algorithm-specific operation results.
| ā | Before feature selection | After feature selection | Precision | Recall | F1 | Precision | Recall | F1 |
| Good | 88.09 | 83.56 | 85.76 | 88.03 | 83.47 | 85.69 |
| Good Sink | 61.27 | 52.91 | 56.78 | 61.27 | 52.91 | 56.78 |
| Good Source | 53.13 | 25.37 | 34.34 | 53.13 | 25.37 | 34.34 |
| Good Auxiliary | 77.16 | 96.34 | 85.69 | 77.16 | 96.34 | 85.69 |
| Bad | 77.08 | 49.23 | 60.09 | 77.08 | 49.23 | 60.09 |
| Bad Sink | 67.94 | 77.47 | 72.4 | 67.94 | 77.47 | 72.39 |
| Bad Source | 90.71 | 96 | 93.23 | 90.66 | 95.87 | 93.12 |
| Average | 73.63 | 68.7 | 69.76 | 73.61 | 68.67 | 71.05 |
| Accuracy | 82.55 | 82.53 |
| Time | 17.03 | 15.94 |
|
|