2. WHAT ARE WE LEARNING ?
What is a
recommender
system?
Demo of our
recommender
system
Evaluation of a
recommender
system
3. WHAT IS A RECOMMENDER
SYSTEM?
Recommender systems aim to
predict users’ interests and
recommend product items that
quite likely are interesting for
them.
They are among the most
powerful machine learning
systems that online retailers
implement in order to drive sales
and understand customer
behavior.
4.
5. AIM OF OUR MODEL
To build a recommender
system that
recommends top N
movies to a user.
6. DATA WE HAVE USED
Movielens dataset available on the internet.
• MovieLens data sets were collected by the GroupLens Research Project
at the University of Minnesota.
• This data set consists of:
* 100,000 ratings (1-5) from 943 users on 1682 movies.
* Each user has rated at least 20 movies.
* Simple demographic info for the users (age, gender, occupation, zip)
7. TYPES OF SIMILARITIES IN RECOMMENDER
SYSTEMS
ITEM SIMILARITIES
Finding items who are similar
to a particular item and
recommending those.
Good for new users.
Items tend to be permanent
USER SIMILARITIES
Finding Users whose interest
or likings may be closer to the
user we want to recommend
the system for
User choices keep changing
9. OTHER ASPECTS TO LOOK AT
Other than just considering the user or items similar to a
particular item, its better to add more details in our Top N
recommendations such as
-> Ratings of a movie
-> Popularity of a movie
-> Similarity threshold