Table 10 Results of the McNemar’s test conducted on MobileNet_v3_large and Efficientnet_b0 transfer learning models: In every situation, the null hypothesis is rejected.

From: Wrapper-based deep feature optimization for activity recognition in the wearable sensor networks of healthcare systems

McNemar’s Test

p-value

Model compared with

HARTH14

KU-HAR15

HuGaDB16

MobileNet_v3_large43

0.0395

0.0287

0.0368

Efficientnet_b033

0.0426

0.0343

0.0156

Back to article page