The KMeans Algorithm

1
The K-Means Algorithm

• Choose a value for K, the total number of
  clusters to be determined.
• Choose K instances (data points) within the
  dataset at random. These are the initial cluster
  center.
• Use simple Euclidean distance to assign the
  remaining instances to their closest cluster
  center.
• Use the instances in each cluster to calculate a
  new mean for each cluster.
• If the new mean values are identical to the mean
  values of the previous iteration the process
  terminates. Otherwise, use the new means as
  cluster centers and repeat steps 3-5

2

• The optimal clustering for the K-means algorithm
  is frequently defined as a clustering that shows
  a minimum summation of squared error differences
  between the instances and their corresponding
  cluster center.
• Finding a globally optimal clustering for a given
  value of K is nearly impossible as we must repeat
  the algorithm with alternative choices for the
  initial centers.
• Choose a terminating criterion such as a maximum
  acceptable squared error value, and execute the
  K-means algorithm until we achieve a result that
  satisfies the termination condition.