Table 4.

Model performance in each measure (Experiment 2).

Condition	Accuracy	Macro average			Weighted average
		Mean precision	Mean recall	Mean f1 score	Mean precision	Mean recall	Mean f1 score
Responses in Experiment 1	0.800	0.690	0.620	0.620	0.770	0.800	0.780
Classification criteria	0.570	0.510	0.520	0.500	0.640	0.570	0.600
Combination	0.790	0.670	0.650	0.660	0.780	0.790	0.780

The highest performances in each performance measure are provided in bold.