In this project I have created different recommendation engine using similarity, collaborative filtering, content based filtering and mixed factorization.

These notebooks runs on anaconda environment and the requirements of this environment can be installed using the file “requiremnt.txt”

Data Was downloaded from MovieTweetings Data.

• movies - a dataframe of all of the movies in the dataset along with other content related information about the movies (genre and date)
• reviews - this was the main dataframe used before for collaborative filtering, as it contains all of the interactions between users and movies.
• all_recs - a dictionary where each key is a user, and the value is a list of movie recommendations based on collaborative filtering

• Simple Recommentation Engine.ipynb : Will Recommend n top most popular movies with filters of year of release and genre
• Collaborative Filtering .ipynb : In collaborative filtering, collaboration of user-item recommendations is used to assist in making new recommendations
• Content Based Recommendations .ipynb : Will recommend movies using content based filtering

Credit for this project goes to Udacity. This project is a part of Udacity Data Science Nanodegree program.