ML_Foundatations_Part2_Introduction to Unsupervised Learning

Unsupervised Learning : K-Means, PCA & KDE

This mini-project demonstrates the power of unsupervised learning through two core algorithms:

• PCA (Principal Component Analysis) for dimensionality reduction
• KDE (Kernel Density Estimation) for probabilistic density modeling

Both techniques are implemented in Python using scikit-learn and matplotlib, with visual comparisons of raw vs. transformed data.

---

📂 What's Inside?

PCA – Dimensionality Reduction

• Input: Synthetic 3D data (make_blobs)
• Goal: Reduce from 3D to 2D using PCA
• Output: Clear 2D visualization of variance-preserving projection

KDE – Density Estimation

• Input: 1D data from two distributions
• Goal: Estimate smoothed probability density
• Output: Histogram vs. KDE curve comparison

---

Getting Started

Prerequisites

Make sure the following libraries are installed:

pip install numpy matplotlib scikit-learn