sweethome

---

California House Price Prediction Project

This repository contains the code and resources for a project focused on predicting house prices in California based on census data from 1990. The goal of this project was to create predictive models using machine learning algorithms to estimate median house values within a block.

Project Overview:

• Context: The dataset used in this project is derived from the second chapter of Aurélien Géron's book, 'Hands-On Machine learning with Scikit-Learn and TensorFlow.'
• Dataset: It contains information about houses in various California districts, including geographic coordinates, housing age, room counts, population, median income, and proximity to the ocean.
• Key Features:
  • longitude, latitude, housing_median_age, total_rooms, total_bedrooms, population, households, median_income, ocean_proximity.
• Goal: To predict the median_house_value based on the provided features using machine learning models.

Project Structure:

• Jupyter Notebook: Includes exploratory data analysis, data cleaning, visualization, feature engineering, and machine learning model implementation.
• Models: Trained models (e.g., Linear Regression, Random Forest Regressor) saved for future use.
• Data: Original dataset used for analysis and modeling.
• Web App: Hosted web application for house price prediction.
• Project Proposal: Initial proposal outlining project objectives, methods, and expected outcomes.

Usage:

1. Jupyter Notebook: Review the notebook for detailed analysis and model implementation.
2. Model Files: Access trained models for further use or enhancement.
3. Web App: Visit the hosted web application to predict house prices based on user inputs.

Contributors: Dissanayake D.J.R, Sathsarani H.E.S

Web App URL: https://sweethome-kvq8.onrender.com/

Feel free to explore the project and use the provided resources for your analysis or predictive modeling tasks related to house price prediction.

Web App Preview

---