Project Description
In this project, I worked with consumer finance data from the US Federal Reserve. I built unsupervised machine learning models to segment households that fear they will be unable to get credit.
Overview
- Compared characteristics across subgroups using side-by-side bar charts
- Built K-Means Clustering models
- Conducted feature selections for clustering based on variance
- Reduced high-dimensional data using Principal Component Analysis (PCA)
Language & Tools
- Python (Pandas, Matplotlib, Seaborn, Scikit-learn, SciPy)
Featured Notebooks
- To view featured notebooks, please send me an email at donald@donaldghazi.com