Machine Learning for Mobile

K-means clustering

The goal of this K-means clustering algorithm is to find K groups in the data, with each group having similar data points. The algorithm works iteratively to assign each data point to one of K groups based on the features that are provided. Data points are clustered based on feature similarity.

The K value is assigned randomly at the beginning of the algorithm and different variations of results could be obtained by altering the K value. Once the algorithm sequence of activities is initiated after the selection of K, as depicted in the following points, we find that there are two major steps that keep repeating, until there is no further scope for changes in the clusters.

The two major steps that get repeated are Step 2 ...