0

I have seen that in Machine Learning, the terms "feature" and "label" are used to refer to what I think of as "independent variable" and "dependent variable" (more synonyms from Wikipedia).

The Wikipedia page describing the term "feature" appears to describe independent variables. This discussion also seems to support the idea they are equivalent.

I would like to know if the terms are equivalent and can be used interchangeably. If they are not, what is the difference?

Historical background of the terms would be especially welcome.

ryan-kwan-do
  • 29
  • 5
  • You can also add explanatory and response variables. As noted in the answer below, they stem from the field used. Some from stats, some from machine learning (CS), and some from econometrics. – jonnybazookatone Nov 25 '17 at 09:16

1 Answers1

0

"Feature" and "independent variable" are different terms for the same thing. "Feature" is more common in machine learning, whereas "independent variable" is more common in statistics.

Some more mostly equivalent terms are "covariate", "predictor", and "regression input". They can be used interchangeably

Dheeraj Joshi
  • 1,522
  • 14
  • 23
  • Thanks @Dheeraj! That was my suspicion. Do you have a source? Putting together something for an academic conference – ryan-kwan-do Nov 28 '17 at 21:58
  • Found it. Looks like it has to do with its origins in AI and in image recognition as discussed briefly [in this lesson](https://courses.edx.org/courses/course-v1:Microsoft+DAT203.1x+4T2017/courseware/44704f9bd54a49d7a8c13614464b3748/15670c4ad58c43b38c8f0e82c0b37e99/?activate_block_id=block-v1%3AMicrosoft%2BDAT203.1x%2B4T2017%2Btype%40sequential%2Bblock%4015670c4ad58c43b38c8f0e82c0b37e99). – ryan-kwan-do Dec 07 '17 at 08:28