Predicting Movie Success from Tweets

1:30 pm - 4:00 pm
Expo Hall/Career Fair

Objective: Inspiration
Audience Level: All
Session Type: Poster

Predicting the success of upcoming movies by analyzing Tweets has received much attention recently. But tweets are noisy and sparse. So predictions using them are not reliable. This paper investigates an approach to improve the accuracy by leveraging a much cleaner movie reviews data, using the technique of transfer learning. Experiments show that this approach improves the accuracy tremendously.


, PhD Candidate, Carnegie Mellon University