Why is different rng(1) rng(2) rng(3) in K means clustering?

i want to know 'why is different rng(1), rng(2), rng(3) in K means clustering?'

Please tell me why,,,

and What is seed? I don't know what you mean after reading the example.

[Walter Roberson]

Which implementation are you looking at? kmeans() from the Statistics and Machine Learning Toolbox does not call upon rng()

[rani Jeong]

K-means Clustering keeps changing its center and results.

Therefore, the rng function produces the same result.

But rng (1) rng (2) rng (3) .... I didn't understand why this number yields different results.

Answer 1

[Hiro Yoshino]

K means clustering algorithm generates the mean vectors corresponding to the number of the cluster you provide.

The point is that these mean vectors are generated randomly, however the randomness cannot be natural but it is synthesized by computational methods, i.e., the randomness is controllable.

The function rng allows you to controll the randomness by specifying a seed.

Have a look - rng function

[rani Jeong]

If so, the resulting value depends on the seed value 1 2 3 4 5 ... What does seed mean?

Answer 2

[Hiro Yoshino]

Yes, you are correct.

Seed is a general terminology in computational sciences -

https://en.wikipedia.org/wiki/Random_seed

[rani Jeong]

Thank you!!!