This document describes a social network based object recommendation system that recommends movies to users by analyzing their Facebook data and movie preferences. It connects to Facebook to fetch a user's liked movies and the IMDB dataset. Movies liked by the user's friends are identified and assigned recommendation scores based on common interests between the user and friend like gender, age and genres. The top scoring movies are recommended and relevance is evaluated based on Facebook suggestions and if the user has watched the recommended movie. Future enhancements could include weighting close friends higher and considering actors and directors.

1. Sriram Patil (201305532)
Nishit Soni (201002026)
Jiten Goyal (201101040)
Information Retrieval and Extraction (CSE474)
International Institute ofTechnology Hyderabad
Online Social Network Based Object
Recommendation System

2. Index
Concept & Architecture
Challenges
Problems with usual techniques
Our approach
Recommendation relevance criteria
Future Scope

3. Concept & Architecture
Social network based object recommendation
Recommending movies
Social Network: Facebook
Dataset: International Movie Database (IMDB)
Website
Login with
Facebook and
fetch users liked
movies
Server
Spark Web
Framework
Jetty Web Server
MySQL
IMDB Dataset

4. Challenges
Two different sources (Facebook and IMDB)
Sparsity
Even active users may have liked well under 5% percent of the
movies.
Scalability
Billions of users and Millions of movies.
Duplication
As movie ids are different. Have to match the movies with
names.

5. Problems with usual techniques
Nearest Neighbour algorithms require computation that
grows with both the number of users and the number of
movies. With billions of users and millions of movies, a
typical web based recommender system running existing
algorithms will suffer serious scalability problems.
Because of sparsity, a recommender system based on nearest
neighbour algorithm may be unable to make any movie
recommendation for a particular user. As, a result the
accuracy of the recommendations may be poor.
Ever growing data and users set.

6. Our approach
Fetch friends with whom the user has atleast some common
movies. If no common movies, then select all the friends.
Get movies liked by the friends.
For each movie, we get a recommendation score.
Parameters considered while assigning score:
Friends which share some likes with the user
Friends of same gender
Friends of same age group
Movies with same genre
Sort the score and return top “n” movies.

7. Recommendation relevance criteria
As the movies are recommended from a lot of friends, it is
little tricky to figure out if the recommendations are
relevant.
We used two criteria
Movies suggested by Facebook.
Our results are comparable to Facebook movie suggestions. And even
better in some cases.
It is a good recommendation if the user has already watched
that movie.

8. Enhancements
There are some more parameters which can be considered
while ranking a movie
Friend list like “Close Friends”, “Relatives”, etc can be given a
little extra weight.
Actors and directors of the movies can be considered when
ranking.
Similar recommendation systems can be extended to
recommend music, books, etc.