Introduction to Unsupervised Machine Learning: Clustering Techniques

1. Introduction to Unsupervised Machine Learning

• Definition:
• Unsupervised learning deals with unlabeled data, finding hidden patterns or intrinsic structures without predefined outcomes.
• Difference from Supervised Learning:
• No target variable or ground truth labels.
• Focuses on data exploration and pattern discovery.
• General Applications:
• Market segmentation, anomaly detection, recommendation systems, and more.

2. K-Means Clustering

2.1. What is K-Means Clustering?

Intuition:

• Partitions data into K distinct, non-overlapping clusters based on feature similarity.

Algorithm Steps:

1. Initialization: Select K initial centroids randomly.
2. Assignment Step: Assign each data point to the nearest centroid.
3. Update Step: Recalculate centroids as the mean of assigned points.
4. Iteration: Repeat steps 2 and 3 until convergence.

2.2. Choosing K: The Elbow Method

Concept:

• Plot the explained variance (or within-cluster sum of squares) against the number of clusters.