A Triplet Multimodel Transfer Learning Network for Speech Disorder Screening of Parkinson’s Disease

Table 3

Results of the eight machine learning methods on the MDVR-KCL dataset.

Methods	Evaluation
	Precision (%)	Recall (%)	F1 score (%)	Acc (%)
DT	65.81	46.95	54.80	68.52
GBDT	82.42	38.01	52.02	71.31
LDA	70.25	41.73	52.36	71.48
KNN	89.70	59.42	71.49	80.73
LightGBM	67.76	90.47	77.49	79.00
LR	96.32	71.68	43.81	54.38
RF	91.71	51.17	65.69	78.27
XGBoost	89.66	58.44	70.76	80.79
TmmNet	100.00	75.26	85.88	90.23

The bold values in Table 3 represents the highest results compared with the machine learning classifiers.