Calculating the similarity between prescriptions to find their new indications based on graph neural network

Chinese Medicine

Table 2 Analysis of the results of the two calculated metrics

	Average value	Maximum value	Minimum value	Standard deviation	Kurtosis	Skewness	P25	P75	P95
Cosine similarity	0.607	0.88	0.25	0.0930	− 0.181	0.012	0.54	0.67	0.77
Jaccard coefficient	0.140	0.73	0	0.0923	3.312	1.356	0.07	0.19	0.32

ISSN: 1749-8546