Developing a portable natural language processing based phenotyping system

Table 9 The classification results for major classes without family history related CUIs

	P-Micro	P-Macro	R-Micro	R-Macro	F-Micro	F-Macro
Intuitive
LR	0.8723	0.6473	0.8723	0.5741	0.8723	0.5970
SVM	0.8732	0.6448	0.8732	0.5780	0.8732	0.5989
DT	0.9339	0.6829	0.9339	0.6392	0.9339	0.6509
RF	0.8559	0.6317	0.8559	0.5623	0.8559	0.5838
Textual
LR	0.8886	0.7854	0.8886	0.7083	0.8886	0.7398
SVM	0.8938	0.7865	0.8938	0.7139	0.8938	0.7439
DT	0.9546	0.8164	0.9546	0.7640	0.9546	0.7855
RF	0.8640	0.7665	0.8640	0.6934	0.8640	0.7233

For each task, the best results are bolded. The underlined results can achieve the top 10 results reported in [20]

ISSN: 1472-6947