K-Mean Clustering

K-Mean Clustering is an unsupervised machine learning algorithm used to group unlabelled data into clusters based on similarity. The algorithm partitions n observations into k clusters, where each observation belongs to the cluster with the nearest mean (cluster centroid).

The algorithm works as follows:

• Initialisation: Randomly select k cluster centroids.
• Assignment: Assign each data point to the nearest centroid, forming k clusters.
• Update: Recalculate the centroids by computing the mean of all data points in each cluster.
• Repeat: Repeat the assignment and update steps until the centroids stabilise or the maximum number of iterations is reached.

The K-Mean Clustering component generates several outputs to help analyse and interpret clustering results:

• A 2D cluster visualisation displaying the data projected onto the first two dimensions, with points coloured according to their cluster membership.
• A Total within-clusters sum of squares curve showing the evolution of cluster compactness for the tested numbers of clusters.
• A Results table containing the coordinates of each point on the first two dimensions together with the cluster assignment for each tested configuration.
• A 3D visualisation displaying the clustered data points in three-dimensional space.

Challenges with K-Mean Clustering

K-Mean algorithm has the following limitations:

• Choosing the right number of clusters (k): Selecting an appropriate number of clusters can be difficult.
• Sensitivity to initial centroids: The final clusters can vary depending on the initial random placement of centroids.
• Assumption of spherical clusters: The algorithm assumes clusters are spherical and similarly sized. This can be a problem when the actual clusters in the data are of different shapes or densities.
• Sensitivity to outliers: Outliers can significantly influence centroid positions and affect clustering quality.