On the development of an information system for monitoring user opinion and its role for the public

Journal of Big Data

Table 10 The classification metrics for the undersampled Kazakh texts

Classifier	NB	SVM	LR	k-NN	DT	RF	XGBoost	Average
Accuracy	0.53	0.58	0.72	0.63	0.70	0.74	0.67	0.65
Precision-macro	0.66	0.61	0.72	0.63	0.72	0.76	0.67	0.68
Precision-micro	0.53	0.58	0.72	0.63	0.70	0.74	0.67	0.65
Precision-weighted	0.67	0.61	0.72	0.63	0.71	0.76	0.67	0.68
Recall-macro	0.54	0.58	0.72	0.63	0.70	0.74	0.67	0.65
Recall-micro	0.53	0.58	0.72	0.63	0.70	0.74	0.67	0.65
Recall-weighted	0.53	0.58	0.72	0.63	0.70	0.74	0.67	0.65
F1-score-macro	0.51	0.54	0.71	0.62	0.69	0.73	0.67	0.64
F1-score-micro	0.53	0.58	0.72	0.63	0.70	0.74	0.67	0.65
F1-score-weighted	0.51	0.54	0.71	0.62	0.69	0.73	0.67	0.64
Average	0.55	0.58	0.72	0.63	0.70	0.74	0.67