Self-Adaptive -Means Based on a Covering Algorithm

Algorithm 1

(-means algorithm).

Input:

Output: A set of clusters C₁, C₂,…

 Begin

1: ← sample points uniformly at random from dataset

2: C_new ← C, C_old←

3:   while ∣C_new - C_old∣ ≤ θdo:

4:   C_old ← C_ new

5:   calculate all of the distances between and C_old_j:

         get_distance (, C_old_j), , C_old

6:   assign to the nearest C_old_j

7:   calculate new centroid C_new:

8:   end while

 End