Developing and Movie Recommendation System in R
- 1. DEVELOPING A MOVIE RECOMMENDATION SYSTEM IN R FINAL PROJECT - STATS 135 - JODY SCHECHTER
- 2. OUTLINE • • • • • • Data Overview Exploratory data analysis Overview of model Steps in R Star Wards Example Demonstration R Tricks
- 5. Positive Correlation between the number of reviews and the average rating, as expected.
- 8. OUTLINE OF STEPS IN R (CONT) • Regularized the scores, reducing variance • Given a movie, found highest similarity scores with other movies • • Also: Created functions that solicit and act on user input Also: Found highest similarity scores for movies in same genre
- 9. Average user review largely unchanged by data trimming.
- 10. Most movie pairs have fewer than 50 common reviewers. More movies have 10-19 than 0-9 common reviewers.
- 15. WORKS CITED Conway, Drew, and John Myles. White. Machine Learning for Hackers. Sebastopol, CA: O'Reilly Media, 2012. Print. Do We Really Need Chocolate Recommendations? 18 Oct. 2013. AC209 Homework Set #4. Harvard University, Cambridge, MA. R Development Core Team (2011). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org/. "Recommendation System in R." Yhat. N.p., 19 June 2013. Web. 03 Dec. 2013. <http://blog.yhathq.com/posts/recommender-system-inr.html>.