This repository hosts the code and documentation for a comprehensive capstone project focusing on the application of data science techniques within the real estate domain. The project encompasses data collection, cleaning, exploratory analysis, modeling, recommendation systems, and the deployment of a user-friendly application.
- Data Gathering: Scraping real estate data from various websites including 99acres.
- Data Cleaning and Merging: Preparing the dataset by handling missing values and ensuring consistency before merging.
- Feature Engineering: Introducing new features to enrich the dataset, such as additional room indicators, area specifications, and possession age.
- Exploratory Data Analysis (EDA): Uncovering patterns and relationships within the data through univariate and multivariate analyses.
- Outlier Detection, Missing Value Imputation: Identifying and addressing outliers and missing values using appropriate techniques.
- Feature Selection: Employing various techniques to identify the most impactful variables for modeling.
- Model Selection & Productionalization: Comparing regression models for predicting property prices and deploying the chosen model using Streamlit.
- Building the Analytics Module: Developing visual representations of key insights about the real estate data.
- Building the Recommender System: Creating recommendation models and a user-friendly interface for personalized recommendations.
- Deploying the Application on AWS: Ensuring scalability and accessibility by deploying the application on Amazon Web Services.
The project evaluates various regression models for predicting property prices, including Linear Regression, Support Vector Regression (SVR), Random Forest Regressor, Multi-layer Perceptron (MLP), LASSO Regression, Ridge Regression, Gradient Boosting Regressor, Decision Tree Regressor, K-Nearest Neighbors Regressor, and ElasticNet Regression.
- Clone the repository to your local machine.
- Install the required dependencies specified in the
requirements.txt
file. - Run the main application file to launch the user interface and explore features, predictions, and recommendations.
The application is deployed on Amazon Web Services (AWS) to ensure scalability and accessibility. You can access the deployed application here.
- Pratheek Bedre
Special thanks to Nitish Singh - CampusX for guidance and support throughout the project.
Feedback and contributions are welcome! If you encounter any issues or have suggestions for improvement, please open an issue or submit a pull request.