I have two list of tuples, first list contains Predicted entities and its assertion and second list-CodePudding

I want to find the accuracy of prediction using python.

List1=[('positive rapid hiv test', 'PRESENT'),
 ('liver function tests', 'PRESENT'),
 ('low albumin', 'PRESENT'),
 ('chest x - ray', 'POSSIBLE'),
 ('diffuse nodular opacities bilaterally', 'PRESENT'),
 ('an electrocardiogram', 'PRESENT'),
 ('sinus rhythm', 'PRESENT'),
 ('hiv', 'PRESENT')]

List2=[('positive rapid hiv test', 'PRESENT'),
 ('liver function tests', 'PRESENT'),
 ('low albumin', 'PRESENT'),
 ('chest x - ray', 'PRESENT'),
 ('diffuse nodular opacities bilaterally', 'PRESENT'),
 ('an electrocardiogram', 'PRESENT'),
 ('sinus rhythm', 'POSSIBLE'),
 ('hiv', 'ABSENT')]

Number of tuples may vary, in both list1 and list2

CodePudding user response：

If you assume PRESENT as 1 and ABSENT as 0, you can create a list of predicted and true values with 0 and 1 and use classification_report from sklearn library.

from sklearn.metrics import classification_report

list1 = [1,1,1,1,1,1,1,1]
lsit2 = [1,1,1,1,1,1,1,0]
print(classification_report(y_true=lsit2, y_pred=list1))