Human Gait Analysis: A Sequential Framework of Lightweight Deep Learning and Improved Moth-Flame Optimization Algorithm

Table 3

Confidence interval-based analysis of proposed framework on the CASIA B dataset.

Confidence level	Margin of error
Normal walk
68.3%,	96.005 ± 0.626 (±0.65%)
90%, 1.645	96.005 ± 1.029 (±1.07%)
95%, 1.960	96.005 ± 1.227 (±1.28%)
99%, 2.576	96.005 ± 1.612 (±1.68%)
Walk with a bag
68.3%,	92.59 ± 0.346 (±0.37%)
90%, 1.645	92.59 ± 0.57 (±0.62%)
95%, 1.960	92.59 ± 0.679 (±0.73%)
99%, 2.576	92.59 ± 0.893 (±0.96%)
Walk with a coat
68.3%,	82.53 ± 0.799 (±0.97%)
90%, 1.645	82.53 ± 1.314 (±1.59%)
95%, 1.960	82.53 ± 1.566 (±1.90%)
99%, 2.576	82.53 ± 2.058 (±2.49%)