Table 7.

Comparison of experimental results between the model proposed in this paper and the reference model on the eRisk2017 dataset

Method	precision	recall	F1
UNSLA	0.48	0.79	0.59
FHDO-BCSGA	0.61	0.67	0.64
FHDO-BCSGB	0.69	0.46	0.55
GPLC	0.42	0.50	0.46
TVT-NB	0.42	0.73	0.54
TVT-RF	0.54	0.58	0.56
SGL-CNN	0.56	0.59	0.57
MGL-CNN	0.63	0.57	0.60