Real-Time Remote Patient Monitoring and Alarming System for Noncommunicable Lifestyle Diseases

Table 3

The choice of hyperparameters for each model [22, 23].

Machine learning models	Hyperparameters	Optimum values
Random forest [23]	The depth of the tree (), number of tree models ()	,
Logistic regression [22]	Confidence factor used for pruning (), class weight adjustment (class weight), maximum iteration (max_iter)	, class , ,
Decision tree [22]	Confidence factor used for pruning (), minimum number of instances of each leaf ()	,
K-nearest neighbors [22]	Number of neighbors (), weight function used in prediction (weights)	,
Support vector machine [22]	Confidence factor used for pruning (), kernel type (kernel); maximum iteration(max_iter)	, ,
XGBoost [22]	Depth of the tree (), learning rate, number of estimators, gamma, and several tuning parameters	, ,