Overview / Usage

The problem that we address in this project can be formulated as follows. Let R be the ratings matrix with dimensions (num_users × num_items). The entry rij in the ratings matrix R contains a nonzero rating value given by the user i for the item j. The matrix R is sparse in nature i.e. most of the entries rij are missing. We generally have a few ratings or some purchase history for each user and similarly each item will have been rated by a few users, but most of the entries in the ratings matrix are generally missing. The task at hand is to predict the missing entries in the ratings matrix R. The data that constitutes the ratings matrix R can be collected either explicitly by asking the users to rate the items or by implicitly deriving the ratings for items based on measures such as whether the user purchased the item, or whether the user clicked a certain page and like wise. I have worked on “movie_dataset.csv”, which contains explicit ratings given by users to movies.

Methodology / Approach

Content Based Recommendations: Content Based Recommendation algorithm takes into account the likes and dislikes of the user and generates a User Profile. For generating a user profile, we take into account the item profiles( vector describing an item) and their corresponding user rating. The user profile is the weighted sum of the item profiles with weights being the ratings user rated. Once the user profile is generated, we calculate the similarity of the user profile with all the items in the dataset, which is calculated using cosine similarity between the user profile and item profile. Advantages of Content Based approach is that data of other users is not required and the recommender engine can recommend new items which are not rated currently, but the recommender algorithm doesn’t recommend the items outside the category of items the user has rated.

Technologies Used

Pandas

scikit

numpy

data science

Repository

https://github.com/Spartan-119/Movie_Recommendation_Engine

Other links

• blog