🌲 Forest Cover Type Classification (UCI Covertype Dataset)

This project is part of a Machine Learning internship task (Level 2 - Task 3). The goal is to predict the forest cover type based on cartographic and environmental data using supervised multi-class classification.

📂 Dataset

• Source: UCI Covertype Dataset
• Records: 581,012
• Features: 54 (10 numerical, 44 binary)
• Target: Cover_Type (7 forest cover types)

📌 Objectives

• Clean and preprocess the dataset
• Train and evaluate:
  • Random Forest
  • XGBoost (with feature importance)
• Tune hyperparameters (Bonus ✅)
• Visualize confusion matrices and feature importances

---

🛠️ Tools & Libraries

• Python 3.x
• pandas, numpy
• matplotlib, seaborn
• scikit-learn
• xgboost

---

📊 Model Performance

• Evaluated using classification report and confusion matrix
• Hyperparameter tuning done using GridSearchCV

---

📈 Bonus Tasks Completed

• ✅ Compared Random Forest vs. XGBoost
• ✅ Hyperparameter tuning with GridSearchCV
• ✅ Visualized top 15 important features from XGBoost

---

🚀 How to Run

1. Download the dataset from UCI and extract covtype.data.gz
2. Place it in your working directory
3. Install dependencies