I am using Naive Bayes to classify a set of documents and I was wondering if it possible to get concrete examples of where the classifier does well/fails on the test set. I am measuring accuracy, precision and recall, but in my analysis I can only suppose which examples were good/bad without knowing exactly that it truly happened like that. Is there a way I can see exactly where it works and where not? Thanks!
Asked
Active
Viewed 48 times
1 Answers
0
Simply loop over your test data and check when the prediction is different from the true label.
lejlot
- 64,777
- 8
- 131
- 164