This project is a web-based application that predicts housing prices in Melbourne based on various property features. It includes data visualization, a user-friendly input form, and a prediction model using Support Vector Regression (SVR) to estimate property prices.
Frontend: A React application for the user interface, form inputs, and interactive data visualizations. Backend: A FastAPI application to handle prediction requests, integrate the AI model, and provide a REST API for the frontend. AI Model: An SVR model to process and predict housing prices based on user inputs.
- Download the Project File System
- Install Node.js
- If havent already, include the node_modules folder in the frontend directory.
- Using pip and npm commands, install python, material ui, d3.js, chart.js, fastapi, uvicorn, scikit-learn, joblib, pandas, numpy
- If there are anymore libraries missing, please install them as recommended
- Navigate to the backend directory and type
python model.py. This will train the model and generate simple_model.pkl and scaler.pkl if havent already and set up AI model integration. - Run
uvicorn main:app --reload. This will run the backend server. - Navigate to the frontend directory and type
npm start. This will start the website.
- In the Home page, you can interact with the charts.
- In the Predict page, you can submit data to predict a housing price
- In the About page, you can read about how we made this project.
- Ashaen Manuel
- Tri Nguyen
- Disen Chandula