Manipulated data to recommend products and improve the company's revenue with 90% accuracy. Optimized clustering of user purchases with best feature combinations based on silhouette score. • Used PySpark, Pandas, Seaborn, and Scikit-learn to implement random forest, logistic regression, decision tree and alternating least squares.
Jupyter Notebook
Manipulated data to recommend products and improve the company's revenue with 90% accuracy.
Optimized clustering of user purchases with best feature combinations based on silhouette score.
Used PySpark, Pandas, Seaborn, and Scikit-learn to implement random forest, logistic regression, decision tree and alternating least squares.