computerphysicslab / gocrawler Goto Github PK

On NER with prose to identify people (PERSON) and geographical/political Entities (GPE)
using a 519,369 word corpus about Coronavirus, most frequent resulting entities are:

COVID-19 :: GPE 1036
SARS-CoV-2 :: GPE 774
China :: GPE 332
Hard Nonporous :: PERSON 190
EUA :: GPE 184
Wuhan :: GPE 181
ICU :: GPE 156
Health :: GPE 153
Covid-19 :: LOCATION 151
SARS-CoV-2 | | | | :: GPE 132
Institutional :: GPE 130
Patients :: PERSON 112
Blue Cross :: PERSON 106
Clinical :: PERSON 105
Policy :: GPE 94
Italy :: GPE 93
RT-PCR :: GPE 91
Children :: GPE 91
University :: ORGANIZATION 88
United States :: GPE 87
US :: GPE 85
Blue Shield :: PERSON 85
ECMO :: GPE 79
ACE2 :: GPE 79
CD8 T :: PERSON 78
SARS :: PERSON 74
Disease-2019 :: ORGANIZATION 74
U.S. :: GPE 72
FDA :: GPE 70
PCR :: GPE 68
Medicare Advantage :: PERSON 66
Disease Control :: PERSON 65

https://github.com/jdkato/prose/issues/65