Question

What is set.seed () in R? why is it used in regression and k-means clustering?

Answer 1

Let me explain in simple words,

set seed (value) where value specifies the initial value of the random number seed.

Syntax: set.seed(123)

In the above line,123 is set as the random number value.

The main point of using the seed is to be able to reproduce a particular sequence of 'random' numbers. and sed(n) reproduces random numbers results by seed.

For more information for set.seed() read below pdf of few pages that explains all about set.seed() in detail.

https://www.stata.com/manuals13/rsetseed.pdf

When the k-means clustering algorithm runs, set.seed() is used a randomly generated seed to determine the starting centroids of the clusters.

k-means-clustering-effect-of-random-seed

Comments

Thanks for the info,  can you provide example with small data set?

---

Hi@Sarfaraz,

Please go through the above-given link. You will get some good examples.