Research on Word Vector Training Method Based on Improved Skip-Gram Algorithm

Table 2

Performance of each model on Chinese-text classification tasks.

Model	AVG	CNN	RCNN
Random word vector	80.55	93.50	94.80
Skip gram	85.51	94.02	95.16
	85.99	94.19	95.21
SEING	85.66	94.24	95.32
Proposed	86.33	95.03	95.43