Tell me more ×
Cross Validated is a question and answer site for statisticians, data analysts, data miners and data visualization experts. It's 100% free, no registration required.

I am actually not sure about the recall = 0 and recall = 1. I am trying to draw the recall vs precision curve. Interpolation is taking the max precision between two standard recall level. I am explaining- when I am drawing the recall vs precision curve,

max precision between 0.0 ~ 0.1 will be drawn at recall 0.
max precision between 0.1 ~ 0.2 will be drawn at recall 0.1.
max precision between 0.2 ~ 0.3 will be drawn at recall 0.2.
...
max precision between 0.9 ~ 1.0 will be drawn at recall 0.9.

then what will be the value at recall 1.0?

If I started assigning precisions from recall 0.1 then what will be the precision at recall 0?

In some cases I have seen at recall 0 precision is 1 but did not get any explanation. For average precision its not 1. Not getting any explanation. interpolated 11 standard recall precision curve

It is a interpolated recall-precision curve. What will be the precision at recall 1? I have 11 recall value in x-axis but 10 precision value for y-axis. so, how to draw the curve?

share|improve this question
It is hard to tell what you are asking. The description sounds like you are trying to extrapolate a continuous curve, given by ten points (0,y0), (0.1,y1), ..., (0.9,y9), to the value (1,y). But that's such a routine (and purely mathematical) exercise I figure you must have something else in mind. Do you perhaps have a specific example to show? Your last paragraph is also mysterious, because "not get any explanation" seems to refer to some past conversation which you haven't disclosed here. – whuber Sep 27 '11 at 14:47
Please, register your account so that you can update your question yourself without asking for edit approval. – chl Sep 28 '11 at 6:53

closed as not a real question by whuber May 25 '12 at 18:37

It's difficult to tell what is being asked here. This question is ambiguous, vague, incomplete, overly broad, or rhetorical and cannot be reasonably answered in its current form. For help clarifying this question so that it can be reopened, see the FAQ.

1 Answer

The simplest thing would be to use linear extrapolation from the previous two points.

But you really have limited idea from the data, and one can do the extrapolation by eye easily enough. If you aren't going to really use the value, but are just trying to complete the figure, I would just leave it out and do the plot within the range of the data, as you've done.

share|improve this answer

Not the answer you're looking for? Browse other questions tagged or ask your own question.