Research Article
Mal-Netminer: Malware Classification Approach Based on Social Network Analysis of System Call Graph
Table 8
The importance of feature attributes based on information gain.
| Feature attribute | Dataset_1 | Dataset_2 | Dataset_3 |
| Average weighted in-degree centrality | 1.450 | 0.997 | 1.393 | Average in-degree centrality | 1.441 | 0.974 | 1.384 | Portion of calls with in-degree of 1 | 1.371 | 1.275 | 1.380 | Portion of calls with out-degree of 1 | 1.360 | 0.946 | 1.369 | Portion of calls with weighted out-degree of 1 | 1.107 | 0.675 | 1.302 | Portion of calls with weighted in-degree of 1 | 1.068 | 0.744 | 1.337 | Average distance | 1.015 | 0.430 | 1.369 |
|
|
Dataset_1: adware + benign, Dataset_2: Trojan + benign, and Dataset_3: worm + benign.
|