Table 3.

Physical activity recognition accuracy comparison of the proposed method with other state-of-the-art methods over IM-WSHA inertial data.

Methods	Algorithm Details	Recognition Accuracy of IM-WSHA
Yang et al. [51]	Statistical features fused with multilayer feedforward neural networks	73.27%
Bonomi et al. [52]	Classification with decision trees	78.19%
Attal et al. [53]	Time and frequency domain features wrapped with Hidden Markov Model (HMM) classifier	80.37%
Proposed Work	Statistical, transform, acoustic and frequency features fused with reweighted genetic algorithm	81.92%

Bold letters for Proposed Recognition Accuracy.