Research Article
Selecting Critical Data Flows in Android Applications for Abnormal Behavior Detection
Table 3
Performance comparison of precategory outlier detection on dataset
.
| Category | ā | ā | AUC of all flows | AUC of critical flows |
| LOCATION_INFORMATION | 1168 | 6410 | 0.9028 | 0.9338 | NETWORK_INFORMATION | 2254 | 14086 | 0.9414 | 0.9598 | DATABASE_INFORMATION | 1525 | 7648 | 0.9508 | 0.9689 | UNIQUE_IDENTIFIER | 972 | 11456 | 0.9360 | 0.9751 | CONTENT_RESOLVER | 1067 | 3771 | 0.9494 | 0.9632 | NO_SENSITIVE_SOURCE | 2790 | 14261 | 0.9323 | 0.9557 | CALENDAR_INFORMATION | 1275 | 6928 | 0.9323 | 0.9557 | BLUETOOTH_INFORMATION | 183 | 283 | 0.9643 | 0.9651 | ACCOUNT_INFORMATION | 320 | 542 | 0.5138 | 0.5340 | FILE_INFORMATION | 563 | 1358 | 0.9338 | 0.9506 |
|
|