🧠 Brain Tumor Classification with k-NN and Decision Tree

This project classifies brain tumors using MRI images and the machine learning algorithm: k-Nearest Neighbors (k-NN). It involves preprocessing image data, training models, and evaluating performance using accuracy, precision, recall, and F1-score.

---

📁 Dataset

• Source: Kaggle - Brain Tumor MRI Dataset
• Classes:
  • Glioma Tumor
  • Meningioma Tumor
  • Pituitary Tumor
  • No Tumor

---

⚙️ Workflow

1. Preprocessing:

   • Resize images to 256x256 using OpenCV
   • Flatten and label images

2. Model Training:

   • k-NN: Tested with k = 3, 5, 7, 9, 11
   • Decision Tree: Used max_depth variations

3. Evaluation:

   • Confusion matrix per class (one-vs-all)
   • Metrics: Accuracy, Precision, Recall, F1-Score
   • Visualized using Seaborn heatmaps