📂 KNN Classification (From Scratch)

📝 Description

This notebook demonstrates the mathematical implementation of K-Nearest Neighbors (KNN) without relying on pre-built machine learning model classes. It builds the algorithm logic entirely using NumPy to understand the underlying mechanics of distance-based classification.

🔧 Key Features

• Custom Algorithm: Implements a KNN_Scratch class with:
  • Euclidean Distance calculation (sqrt(sum(diff^2))).
  • Neighbor sorting and Majority Voting.
• Manual Metrics: explicitly calculates True Positives (TP), False Positives (FP), True Negatives (TN), and False Negatives (FN) to derive Precision, Recall, Specificity, and NPV.
• Visualization: Plots the manually generated Confusion Matrix using Seaborn.

📦 Dependencies

• pandas (for data loading)
• numpy (for matrix operations)
• matplotlib & seaborn (for plotting graphs only)