Working on the House Price Prediction Model has been an invaluable learning experience. Using a dataset sourced from Kaggle, I gained hands-on practice in data preprocessing. I meticulously cleaned the data, removed outliers, and ensured data quality, which played a crucial role in preparing the dataset for accurate model training.
This project deepened my understanding of regression and various mathematical concepts like Absolute Error and Root Mean Square Error. I also explored different techniques to enhance model accuracy.
Through extensive experimentation, I identified XGBRegressor as the best-performing model, achieving an impressive 90.50% accuracy. This process reinforced the importance of model selection and hyperparameter tuning in optimizing performance.
Training the model, evaluating different algorithms, and analyzing mathematical aspects were both challenging and rewarding. I spent a significant amount of time refining this project, and I’m still learning and improving.
To make the model accessible, I built a web app using Streamlit, which provided an easy way to deploy and share my work. While I have basic knowledge of platforms like AWS and Heroku, I chose Streamlit.io for its simplicity and free hosting. Deploying was straightforward after pushing the project to GitHub, I deployed it seamlessly through Streamlit.
📂 View the Source Code
Click here to view the full code