Objective: Built a recommendation system for Amazon using collaborative filtering, ranking, and matrix factorization to enhance customer satisfaction and product discovery.

Approach: Utilized models such as User-User and Item-Item similarity, along with an optimized SVD matrix factorization model to recommend products based on user ratings and preferences.

Skills and Tools: Python, TensorFlow, Scikit-learn, Pandas, EDA with Matplotlib, Seaborn, and model tuning for optimal accuracy and recall.

After running and tuning models, I recommended the SVD model for Amazon due to its superior precision and accuracy, enhancing customer satisfaction and reinforcing Amazon’s position as a go-to platform for product discovery.