A machine learning project using regression analysis to predict housing prices in the Boston area. Built using scikit-learn, this model analyzed features like number of rooms, location, and school quality to generate price predictions.
The project was part of a broader study into data preprocessing, model selection, and validation strategies using real-world datasets.
Tools Used:
Python, Pandas, scikit-learn, Matplotlib
Google Collaborator, CSV-based datasets
Feature scaling and normalization techniques
Category:
Client:
Project Duration:
Data Pipeline & Model Training
Cleaned raw dataset (missing values, outliers, normalization)
Applied one-hot encoding and feature scaling for categorical variables
Split data into train-test subsets with cross-validation
Trained a linear regression model with R² of 0.85
Visualized performance using scatter plots and residual analysis
Tuned parameters and evaluated MAE, RMSE for model accuracy
Generated feature importance ranking and insights for real estate decision-making
Demonstrated analytical thinking, reproducibility, and performance documentation
