I want to calculate precision and recall for word sense disambiguation. Naturally, for that I need to calculate TP, TN, FP and FN. I know TP is the number of documents that the tag of the test sentences equals to the tag of the classifier and FN is the number of documents that the tag of test sentences is not equal to the tags that the classifier found. But what is TN and FP? I can't see any other case more than the two cases I described above. For example my test sentence is as like as below:

fist sentence:         word1 word2 word3 word4 tag1
second sentence:  word1 word2 word3 word4 tag2

If my classifier finds the correct tag which is equal to the tag in sentences, it's TP unless it's FN. So what is TN and FP? My classifier finds whether the right tag or the wrong. There is no other case. Is there any TN and FP in this condition?

  • $\begingroup$ I think you will find the information you need in the linked thread. Please read it. If it isn't what you want / you still have a question afterwards, come back here & edit your question to state what you learned & what you still need to know. Then we can provide the information you need without just duplicating material elsewhere that already didn't help you. $\endgroup$ – gung - Reinstate Monica Nov 12 '16 at 1:13