BETA
K-MEANS CLUSTERING ALGORITHM

Author

Alienator Ashraf
Updated
July 17, 2015
8:39 pm
Stats
0 comments
22 ratings
4.83 / 5

K-Means clustering algorithm

K-Means is a relatively simple clusering algorithm for grouping unlabeled dataset.

Steps

  1. Get random points from the dataset
  2. Classify the dataset according to the points
  3. Find centers of the classified data
  4. Goto #2 with the new center points, until
    1. Iteration exceeds the limit
    2. There is no change in the classified labels

Implementation dataset

This algorithm uses the popular Fisher's Iris data set. There are two CSVs in this project, you can edit the code in inputs.js to switch between them and experiment with the results.

Further reading

  1. K-Means clustering
  2. Iris dataset

Source Code



Embed this

Comments
No comments found.
Sign-in to share your thoughts.