I am developing a search algorithm which looks at lines of text and identifies terms of intrest. Once completed, these terms are compared with a gold standard. Sample results look like this:
What I dont understand is how to calculate the predictive performance of my solution. from the above example, I can tell that True positives -> B and FP -> C.
Bu with the above data, we cannot really have a measure of false negative or true negative can we?
if so, we cant calculate any performance metric other than precision?