Table 2 Prescription recommendation testing on closed-set and open-set

From: Plant disease prescription recommendation based on electronic medical records and sentence embedding retrieval

Model	Closed-set (%)			Open-set (%)
Model	TOP.1	TOP.3	TOP.5	TOP.1	TOP.3	TOP.5
BERT-sBERT	79.67	87.87	91.15	75.24	89.81	92.48
roBERTa-sBERT	79.02	87.21	91.15	76.21	90.53	93.45
macBERT-sBERT	78.69	86.89	90.82	76.94	90.29	95.39
BERT-CoSENT	86.23	96.07	98.03	81.07	91.26	94.66
RoBERTa-CoSENT	86.56	95.41	97.05	80.34	92.72	93.45
PRSER(our model)	88.20	96.07	97.70	82.04	91.50	94.90

The bolded values indicate the best-performing results in each column

Back to article page

ISSN: 1746-4811

Contact us

Submission enquiries: journalsubmissions@springernature.com