Skip to main content
Cross Validated

Return to Question

edited title
Link
Tim
Tim
  • 141.2k
  • 26
  • 270
  • 512

When to use dot-product as a similarity metric

Post Reopened by Karolis Koncevičius, JacKeown, Adrian Keister
added 19 characters in body; edited tags
Source Link
JacKeown
JacKeown
  • 728
  • 1
  • 7
  • 20

I'm trying to decide what sort ofunderstand which similarity measure I should usebe used in which situations.

  • Roughly speaking, when should someone use the dot product to assess similarity between vectors?
  • Roughly speaking, when should someone use the norm of the vector difference to assess similarity between vectors?
  • What are the pros and cons of these approaches?

I'm not looking for a specific answer on which one I should use for a specific problem. I'm wondering more generally about when these approaches should be expected to work well.

I'm trying to decide what sort of similarity measure I should use.

  • Roughly speaking, when should someone use the dot product to assess similarity between vectors?
  • Roughly speaking, when should someone use the norm of the vector difference to assess similarity between vectors?
  • What are the pros and cons of these approaches?

I'm not looking for a specific answer on which one I should use for a specific problem. I'm wondering more generally about when these approaches should be expected to work well.

I'm trying to understand which similarity measure should be used in which situations.

  • Roughly speaking, when should someone use the dot product to assess similarity between vectors?
  • Roughly speaking, when should someone use the norm of the vector difference to assess similarity between vectors?
  • What are the pros and cons of these approaches?

I'm not looking for a specific answer on which one I should use for a specific problem. I'm wondering more generally about when these approaches should be expected to work well.

Please un-close this question. I don't think it needs more details, because I'm looking specifically for a general response, not an answer for what to use for a specific problem.
Added to review
Source Link
JacKeown
JacKeown
  • 728
  • 1
  • 7
  • 20

I'm trying to decide what sort of similarity measure I should use.

Roughly speaking, when should I use the dot product? Roughly speaking, when should I use the norm of the vector difference? What are the pros and cons of these approaches (and possible others)?

  • Roughly speaking, when should someone use the dot product to assess similarity between vectors?
  • Roughly speaking, when should someone use the norm of the vector difference to assess similarity between vectors?
  • What are the pros and cons of these approaches?

I'm not looking for a definitivespecific answer, because I know on which one I haven't provided enough context, but I'm just lookingshould use for some general guidance on the mattera specific problem. I'm wondering more generally about when these approaches should be expected to work well.

I'm trying to decide what sort of similarity measure I should use.

Roughly speaking, when should I use the dot product? Roughly speaking, when should I use the norm of the vector difference? What are the pros and cons of these approaches (and possible others)?

I'm not looking for a definitive answer, because I know I haven't provided enough context, but I'm just looking for some general guidance on the matter.

I'm trying to decide what sort of similarity measure I should use.

  • Roughly speaking, when should someone use the dot product to assess similarity between vectors?
  • Roughly speaking, when should someone use the norm of the vector difference to assess similarity between vectors?
  • What are the pros and cons of these approaches?

I'm not looking for a specific answer on which one I should use for a specific problem. I'm wondering more generally about when these approaches should be expected to work well.

Post Closed as "Needs details or clarity" by whuber♦
Source Link
JacKeown
JacKeown
  • 728
  • 1
  • 7
  • 20