RentVision AI is a machine learning–powered web application that predicts residential property rent based on key features such as size, BHK, location, and furnishing details. The system leverages a Random Forest Regressor, providing more robust and accurate predictions by combining multiple decision trees. Users can obtain fast, data-driven rent estimates through an interactive web interface.
Website : https://rentvision.onrender.com
- Predict monthly rent instantly using ML
- Random Forest–based regression model (ensemble learning)
- Data preprocessing pipeline (scaling + encoding)
- Fully functional web app (Flask backend)
- Modern UI with responsive design (HTML, CSS, JS)
- Light/Dark mode toggle
- Mobile-friendly layout
- Deployed on Render
-
Data Collection (Housing dataset)
-
Data Cleaning & Preprocessing
- Handling categorical variables (One-Hot Encoding)
- Feature scaling for numerical inputs
-
Feature Engineering
-
Model Training (Random Forest Regressor)
-
Model Evaluation
- R² Score
- RMSE
-
Model Deployment using Flask
-
Algorithm: Random Forest Regressor
-
Type: Ensemble Learning
-
Advantage:
- Reduces overfitting compared to single decision trees
- Handles non-linear relationships effectively
- Provides more stable predictions
RentVision-AI/
│
├── app.py
├── model.py
├── rent-model.pkl
├── requirements.txt
│
├── templates/
│ └── index.html
│
├── static/
│ ├── style.css
│ └── script.js
│
└── dataset/
└── House_Rent_Dataset.csv
- R² Score: ~0.69
- Evaluation Metric: RMSE
The predicted rent is an estimate based on historical data and may vary depending on additional real-world factors such as exact locality, amenities, and market conditions.
- Add locality-level predictions
- Try Gradient Boosting / XGBoost for higher accuracy
- Add analytics dashboard
- Migrate backend to FastAPI
- React frontend for advanced UI
Affan Khan .