Table 5. Comparison of predictive accuracy across for random forest and SVM models.

Classifier	Number of Plants	Price	F1-macro score (f1_macro)				F1-micro score (Overall Accuracy) (f1_micro/OA)
			Model 1	Model 2	Model 3	Model 4	Model 1	Model 2	Model 3	Model 4
RF	6	Low	0.25	0.32	0.24	0.30	0.72	0.78	0.67	0.72
SVM	6	Low	0.20	0.20	0.20	0.24	0.67	0.67	0.67	0.67
RF	6	High	0.28	0.28	0.28	0.28	0.53	0.53	0.53	0.53
SVM	6	High	0.16	0.16	0.16	0.16	0.47	0.47	0.47	0.47
RF	12	Low	0.29	0.36	0.28	0.29	0.55	0.73	0.55	0.55
SVM	12	Low	0.17	0.40	0.34	0.34	0.36	0.73	0.64	0.64
RF	12	High	0.18	0.26	0.20	0.30	0.47	0.53	0.53	0.60
SVM	12	High	0.10	0.10	0.10	0.10	0.33	0.33	0.33	0.33
RF	24	Low	0.16	0.19	0.11	0.39	0.38	0.38	0.25	0.75
SVM	24	Low	0.18	0.28	0.40	0.38	0.38	0.50	0.75	0.75
RF	24	High	0.33	0.25	0.39	0.19	0.55	0.45	0.73	0.36
SVM	24	High	0.31	0.08	0.14	0.04	0.55	0.18	0.27	0.09

Model predictors:

Model 1: 3mer-without-LATD;

Model 2: 3mer-without-LATD + consumer attributes;

Model 3: 3mer-with-LATD;

Model 4: 3mer-with-LATD + consumer attributes.

Note: The numerically superior result in each display under either F1-macro or F1-micro score. The higher F1-macro score and the higher F1-micro score (is equivalent to overall accuracy in classification tasks) are displayed in bold.