Table 14 Complementary baselines from related security domains.

From: FL-MalDrift: a federated learning framework for malware detection under local concept drift

Method

Domain / Dataset

Acc./F1

Drift

Privacy

Comm.

Hybrid Multilevel12

Android (KronoDroid)

\(\approx\)91.3 (F1)

Full

Moderate

Low.

FEDDBN-IDS22

IDS (AWID)

88–99 acc.

None

High

Low

AI-driven IoMT30

IoMT / PE malware

99.6 acc.

None

Moderate

High

  1. These works achieve strong results in their domains but differ from our setting. IoMT reports high centralized accuracy but lacks federated privacy and drift adaptation
Back to article page