Built an end-to-end hybrid movie recommendation system using User-Based CF, Item-Based CF, and SVD on 1,682 movies and 943 users, achieving 32.8% Precision@10 and 0.39 NDCG, with real-time TMDb API…
Developed an unsupervised ML model to segment mall customers by income and spending behavior, achieving optimal clustering (K=5) and powering a Streamlit dashboard for targeted marketing and customer retention strategies.
Developed a machine learning model to predict student performance with 77% accuracy, delivering actionable insights and interventions through a Streamlit web app.
