[Retracted] Rapid Classification and Analysis for E-Commerce Goods Based on Multitask Learning

Table 5

Comparison of the classification results of each method on the test set.

Measurement indicators	Comparison method	Training set proportion (%)
		1	5	10	20	50
Accuracy	Item2vec	21.3	31.0	34.6	37.8	41.6
	w-item2vec	29.6	42.9	47.5	52.2	57.1
	Random	2.7	2.9	2.6	2.9	3.4
	I-CF	8.5	18.6	24.3	32.9	43.2
	PMF	11.1	15.9	16.9	18.1	19.2
F1-macro	Item2vec	1.0	2.1	2.8	4.1	6.7
	w-item2vec	2.3	6.1	8.6	12.4	20.3
	Random	0.03	0.04	0.04	0.04	0.04
	I-CF	0.73	3.2	5.2	8.6	14.6
	PMF	0.3	0.5	0.6	0.8	1.5