https://house-price-predictor-ana5.streamlit.app/
This project predicts the selling price of a house based on various features such as area, number of bedrooms, bathrooms, location, and amenities.
It is a Supervised Machine Learning Regression Problem because:
The dataset contains labeled values (SalePrice) The output is a continuous numerical value
The goal is to build a reliable model that can estimate property prices accurately for unseen data.
Understand housing dataset
Clean and preprocess data
Engineer useful features
Train multiple regression models
Evaluate performance using proper metrics
Select the best model for prediction
In the real estate industry, accurately determining the price of a house is a challenging task. Property prices are often estimated based on manual judgment, market assumptions, or limited comparisons, which can lead to incorrect pricing decisions.
Without a data-driven pricing system:
Houses may be overpriced or underpriced
Buyers struggle to evaluate fair property value
Real estate agencies face difficulty in decision making
The goal of this project is:
To predict house selling prices based on property characteristics using supervised machine learning techniques.
By building a house price prediction model, stakeholders can:
--> Estimate accurate market value of properties --> Support data-driven real estate decisions --> Help buyers and sellers make informed pricing choices --> Improve efficiency in property valuation
The project uses three files:
| File | Purpose |
|---|---|
| train.csv | Used to train the model |
| test.csv | Used to predict house prices |
Target Column: SalePrice
Features include:
Living Area (square feet)
Bedrooms
Bathrooms
Location
House Age
Garage capavity
Overall Quality
Basement Area(square feet)
The project begins with understanding and exploring the housing dataset using Exploratory Data Analysis to identify important features affecting house prices. Data preprocessing is performed by handling missing values and encoding categorical variables. The dataset is then split into training and testing sets. Linear Regression and Random Forest models are trained to predict house prices, and their performance is evaluated using RMSE and R² score. The best-performing model is selected to generate accurate house price predictions.
Figure: House Price Prediction Block Diagram