Imperial College London researchers unveiled EARLYCROW, a new method of detecting advanced persistent threats (APTs), a sophisticated kind of cyberattack, by identifying malicious network traffic. It uses a new format called PAIRFLOW which captures extensive connection-level data. EARLYCROW, when tested with real-world datasets, registered a macro-average F1-score of 93.02% on unseen APT samples, with a false positive rate of 0.74%.
![](https://healsecurity.com/wp-content/uploads/2025/02/image_1345812496.jpg)
Apple, Google take down apps with malware after Kaspersky’s findings – report
Apple and Google have removed about 20 applications from their app stores due to malware. Security experts at Kaspersky discovered that these apps had been