Document is current
Any future updates will be listed below
-
Authors
Chen, Jiahui
Wu, Mingrui
Gan, Wensheng
Huang, Huiwu
Lau, Terry Shue Chien
-
Funding
Funding for this research was provided by:
Guangdong Provincial Key Laboratory of Power System Network Security (GPKLPSNS-2023-KF-04)
- License Information
-
More Information
Article History
Received: 2 June 2024
Accepted: 26 April 2025
First Online: 26 May 2025
Declarations
:
: The authors declare no competing interests.
: In malware detection, false positives happen when the system mistakenly labels benign software as malicious. This can impact user experience and reduce trust in the system. In businesses, false positives can also burden security teams, as they must check and fix the wrong alerts. Conversely, false negatives occur when malware is wrongly seen as safe, letting it run unchecked and posing real security risks. In the Dumpware10 dataset, the MIL-CNN model shows the fewest false negatives when the memory dump image is set to resolution. At resolution, its false positive rate is similar to that of MobileNetV2-CA. This balance helps the model improve security while reducing user inconvenience, aligning better with ethical standards.
