What does c(i) representing for each training examples?

Can anyone explain me what c(i) is actually representing? and What is its significance?

Hi @Ashutosh_Bharti ,

Please refer to the K-means Algorithm lecture video, Prof Ng has explained the algorithm in details. Below is a screen shot of the first step of the algorithm.

Hi @Ashutosh_Bharti,

‘c’ array maps each training example with its closest centroid ‘K’. The ‘i’ in ‘c(i)’ represents index of ‘ith’ training example in ‘x’ . For example, if the value of ‘c[5]’ is 3, then it represents 5th training example is assigned to the centroid 3 (K=3). it is used to track which centroid each training example assigned to which will helps to peform clustering.

I kindly recommend referring to the video for a more detailed explanation, as @Kic has mentioned earlier.


Thank you Mujassim. Got it