WCL: Client Selection in Federated Learning with a Combination of Model Weight Divergence and Client Training Loss for Internet Traffic Classification

Table 2

The improvement effect of the model accuracy of the WCL algorithm on the QUIC dataset.

Client number	Improvement on QUIC
	High data heterogeneity	Low data heterogeneity
30	35.4%	23.9%
40	39.3%	26.7%
50	31.6%	25.5%