Developing a portable natural language processing based phenotyping system

Table 8 The classification results for major classes on all CUIs corresponding to the original records

	P-Micro	P-Macro	R-Micro	R-Macro	F-Micro	F-Macro
Intuitive
LR	0.8709	0.6457	0.8709	0.5733	0.8709	0.5960
SVM	0.8724	0.6444	0.8724	0.5770	0.8724	0.5981
DT	0.9311	0.6804	0.9311	0.6374	0.9311	0.6488
RF	0.8466	0.6226	0.8466	0.5559	0.8466	0.5765
Textual
LR	0.8882	0.7846	0.8882	0.7085	0.8882	0.7397
SVM	0.8930	0.7858	0.8930	0.7135	0.8930	0.7434
DT	0.9545	0.8167	0.9545	0.7636	0.9545	0.7854
RF	0.8882	0.7846	0.8882	0.7085	0.8882	0.7397

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

ISSN: 1472-6947