Employment of Ensemble Machine Learning Methods for Human Activity Recognition

Table 3

Performance metrics of trained ML models (without tuning).

Algorithm	Accuracy	Precision (micro)	Precision (macro)	Precision (weighted)	Recall (micro)	Recall (macro)	Recall (weighted)	F1-score (micro)	F1-score (macro)	F1-score (weighted)	Cross-validation score
SVC	0.9837	0.9837	0.9837	0.9837	0.9837	0.9837	0.9837	0.9837	0.9837	0.9837	0.9681
LSVC	0.9816	0.9816	0.9815	0.9815	0.9816	0.9816	0.9816	0.9816	0.9815	0.9815	0.9701
NuSVC	0.9387	0.9387	0.9401	0.9401	0.9387	0.9387	0.9387	0.9387	0.9388	0.9388	0.926
AdB	0.4008	0.4008	0.2864	0.2863	0.401	0.4010	0.4008	0.4008	0.2723	0.2721	0.4049
XGB	0.9764	0.9764	0.9763	0.9763	0.976	0.9764	0.9764	0.9764	0.9763	0.9763	0.9547
LGBM	0.9747	0.9747	0.9746	0.9747	0.975	0.9747	0.9747	0.9747	0.9747	0.9747	0.9475
GB	0.9587	0.9587	0.9600	0.9590	0.9689	0.9699	0.9689	0.9587	0.9597	0.9587	0.9148
ETC	0.9627	0.9627	0.9626	0.9626	0.9597	0.9597	0.9597	0.9627	0.9626	0.9626	0.9395