why two vectors is not similarity but result is 1?

Question

I'm using Cosine Similarity formula to caculate similarity between two vectors. I tried two different vectors like this:

Vector1(-1237373741, 27, 1, 1, 331289590, 1818540802)

Vector2(-1237373741, 49, 1, 1, 331289590, 1818540802)

Two vectors has a little different, but the result is 1.

I don't know why? Anyone can explain this problem for me? thanks so much.

Can you show us the code you used to generate that result, as well as what you expect the result to be? — Arya McCarthy, Apr 08 '17 at 20:20
i tried my code and tried programs online at some websites, that is the same result, and i think my code is not wrong. — atimetoremember, Apr 08 '17 at 20:23
I think, if the result equal 1, that mean, two vectors is the same. but two vectors has a litte different, but the result is also 1. — atimetoremember, Apr 08 '17 at 20:25
I bet it's just a roundoff problem, since you're dealing with values on such different magnitudes. Would it benefit you to [scale](https://en.wikipedia.org/wiki/Feature_scaling) your data? — Arya McCarthy, Apr 08 '17 at 20:28
What do you mean by, the result is 1? What calculation is involved? — bestprogrammerintheworld, Apr 08 '17 at 20:35

score 0 · Answer 1 · answered Apr 09 '17 at 17:03

0

For the most part, those two vectors are are pointing in the same direction (The larger coordinates are going to dominate the smaller differences in the other coordinate). A cosine similarity of ~1 is expected (Remember that cos(0) = 1)

answered Apr 09 '17 at 17:03

muzzlator

742
4
10

why two vectors is not similarity but result is 1?

1 Answers1

Linked

Related