mlatoz

K-Means Clustering

Good

• Simple to implement.
• Scales to large datasets.
• Guarantees convergence.
• Easily adapts to new examples.
• Generalizes to clusters of different shapes and sizes.

Bad

• Sensitive to the outliers.
• Choosing the k values manually is tough.
• Dependent on initial values.
• Scalability decreases when dimension increases.

---

What is Clustering? (Supervised vs Unsupervised Learning)

• Clustering can be defined as grouping an unlabelled data

---

The Elbow Method

• Within Cluster Sum of Squares (WCSS):

   WCSS = ∑(Pi in Cluster 1) distance(Pi, C1)2 + ∑(Pi in Cluster 2) distance(Pi, C2)2 + ...

---

K-Means++

• K-Means++ Initialization Algorithm: Step 1: Choose first centroid at random among data points Step 2: For each of the remaining data points compute the distance (D) to the nearest out of already selected centroids Step 3: Choose next centroid among remaining data points using weighted random selection - weighted by D² Step 4: Repeat Steps 2 and 3 until all k centroids have been selected Step 5: Proceed with standard k-means clustering

---

Download Resources

• Python Notebook
• Dataset
• K-Means Clustering | R Code

---

«Previous

Next»