Housing Price Prediction using Linear Regression

This project demonstrates a basic linear regression model to predict housing prices based on various factors such as location, population, and median income. The dataset used is Housing.csv, and the code is implemented in Python.

Project Structure

• main.py: Python script containing the data processing, visualization, and model training.
• Housing.csv: Dataset used for model training and testing.
• README.md: Documentation of the project.

Dataset

The Housing.csv dataset includes features like:

• longitude and latitude: Geographic coordinates.
• housing_median_age: Age of the housing in the area.
• total_rooms and total_bedrooms: Counts of rooms and bedrooms.
• population and households: Population and household counts.
• median_income: Median income in the area.
• median_house_value: Target variable for prediction.
• ocean_proximity: Proximity to the ocean (categorical feature).

Methodology

1. Data Preparation:

   • Stratified sampling is used to ensure consistent income distribution in training and test sets.
   • One-hot encoding converts the categorical ocean_proximity feature to numeric.
   • Missing values in total_bedrooms are filled with the median for accurate model training.

2. Data Visualization:

   • A scatter plot visualizes housing data distribution by population and geographic location.

3. Model Training:

   • We use linear regression to model the relationship between the features and the target (median_house_value).
   • Predictions are generated on a sample to compare with actual values and evaluate model performance.

Results

Sample predictions vs. actual values:

• Predictions: [88983.15, 305351.35, 153334.71, 184302.55, 246840.19]
• Actual values: [72100., 279600., 82700., 112500., 238300.]

How to Run

1. Clone this repository.
2. Install required packages:

   pip install -r requirements.txt

3. Run the script:

   python main.py