Table 6

Results of proposed model on target clients in terms of MAE, MSE, and R2 score.

Algorithm T1
T2
T3
Average
Mean absolute error
FedDNN-full 2.24 2.37 2.81 2.47
FedDNN-par 1.58 2.14 1.34 1.68
FedLSTM-full 2.85 2.01 1.74 2.2
FedLSTM-par 1.47 1.91 1.21 1.53

Mean square error
FedDNN-full 20.14 18.42 21.85 20.13
FedDNN-par 16.32 17.26 18.23 17.27
FedLSTM-full 16.37 19.75 17.34 17.82
FedLSTM-par 12.32 19.68 15.22 15.74

R2 score
FedDNN-full 88.32 84.36 89.20 87.29
FedDNN-par 90.41 85.12 91.23 88.92
FedLSTM-full 87.29 88.24 83.09 86.20
FedLSTM-par 90.23 89.77 84.39 88.15

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.