Movie Rating Prediction System using Content-Boosted Collaborative Filtering

Recommender Systems are becoming a quinessential part of our lives with a plethora of information available and wide variety of choices to choose from in various domains. Recommender sytems have a wide domain of application from movies, books, music to restaurant, financial services etc. Recommender systems apply knowledge discovery techniques to the problem of making product recommendations. In this project we combine the two most common methods of ‘Content-Based’ & ‘Collaborative Filtering’ which works by matching consumer preferences to other costumers in making recommendations. We use the famous dimensionality reduction method of SVD to capture the relationship in users demographics and his/her prefernces. We apply our method on the Movielens data to predict movie ratings for users.