Publications by authors named "Faitouri A Aboaoja"
PeerJ Comput Sci
September 2023
Article Synopsis
- Malware poses a significant security risk, as sophisticated methods make it hard to distinguish between malicious and legitimate software behaviors.
- Existing solutions often use the TF-IDF technique to analyze malware, but this approach inaccurately represents features, leading to high false alarm rates.
- The new Kullback-Liebler Divergence-based Term Frequency-Probability Class Distribution (KLD-based TF-PCD) algorithm improves accuracy by weighing features based on their probability distributions in malware versus benign classes, achieving an impressive accuracy of 0.972.
