In the realm of data analysis and machine learning, clustering stands as one of the fundamental unsupervised learning techniques. Among the various methods, K-means clustering is widely popular due to its simplicity and efficiency. One of the key components when working with K-means is measuring how well each data point fits within its assigned cluster, which is often gauged through measures known as distortions or within-cluster sum of squares (WCSS).

K-Means Clustering Overview

K-means clustering involves partitioning a set of $n$ data points into $k$ clusters, where each point belongs to the cluster with the nearest mean. The steps in K-means include:

1. Initialization: Choose $k$ initial cluster centers randomly.
2. Assignment: Assign each data point to the nearest cluster center.
3. Update: Recalculate the means of the clusters.
4. Repeat steps 2-3 until convergence (i.e., when the assignments no longer change).

Understanding Distortion in K-Means

In K-means, the distortion for each data point is calculated as the Euclidean distance squared between the data point and its nearest cluster center. Distortion is a measure of how well each data point fits within the cluster to which it is assigned.

Formula for Distortion

For a data point $x_i$ assigned to a cluster with center $c_i$, the distortion, $D(x_i)$, can be calculated as:

$$D(x\_i) = || x\_i - c\_i ||^2$$

where $|| . ||$ denotes the Euclidean distance. The total distortion for all data points in the dataset is then the sum of the distortions of individual points.

Calculating Distortions for Each Data Point

When working with a dataset, it's often insightful to calculate and output distortions for each data point. This allows analysts to understand which data points are poorly represented by their clusters and may need reconsideration in clustering criteria or preprocessing steps.

Example

Consider a simple implementation using Python with the popular sklearn library: