Project information

• Machine Learning, University at Buffalo
• Python, SQL, Pandas

• Formulated a model to advise movies to a million users depending on movies watched and ratings.
• Visualized the data and developed a collaborative filter using linear kernel and cosine similarity to find movies watched by similar users developing a relationship of similarity between two users.
• Executed complex SQL queries to filter out movies and users data from the initial raw data to create similarity relation.
• Recommended movies are more than 90% similar to what a single user liked and rated highly