task-3-linear-regression

Linear Regression modeling on House Prices dataset using Python (scikit-learn).

Task 3: Linear Regression on House Prices Dataset

This project applies simple and multiple linear regression techniques using Python and scikit-learn to predict house prices based on numerical features. It is part of my AI & ML internship focused on building predictive modeling skills.

---

📂 Dataset Used

• Name: House Prices – Advanced Regression Techniques
• Source: Kaggle
• File: train.csv
• Target Variable: SalePrice

---

📌 Objective

To predict house sale prices using linear regression models based on relevant features like GrLivArea, OverallQual, and GarageCars.

---

🔍 Key Steps

1. Import & explore dataset
2. Handle missing values (initial check only)
3. Build a Simple Linear Regression model (1 feature)
4. Build a Multiple Linear Regression model (3 features)
5. Evaluate models using:
   • Mean Absolute Error (MAE)
   • Mean Squared Error (MSE)
   • R² Score
6. Visualize the simple regression line

---

🛠️ Tools & Libraries

• Python (Google Colab)
• Pandas, NumPy
• Scikit-learn
• Seaborn & Matplotlib

---

📁 Files in This Repo

File	Description
train.csv	Dataset used for modeling
linear_regression_house_prices.ipynb	Complete notebook with EDA & regression models
README.md	This project overview and documentation

---

✅ Outcome

Both simple and multiple regression models were implemented and evaluated.
The multiple regression model achieved higher accuracy and interpretability.

---

🧠 What I Learned

• Linear regression workflow (fit → predict → evaluate)
• How to interpret model coefficients
• Regression evaluation metrics
• Using visualization to validate predictions

---

📌 Part of a 45-day AI & ML Internship (2025)