Predicting the success of movie based on Social Media.
YouTube is the best platform for movie makers to get the public reaction. It can be useful for them to make appropriate marketing strategy based on the public reaction. The influential of actors and fan following of any actor on Twitter is also an important factor in deciding the success of any movie. In this project, we are using this latest social networking platforms to predict the success of 10 movies. We have selected 10 Hollywood movies to predict the success ration based on their trailer and their actor’s follower count on YouTube. We are getting the comments of YouTube movie trailers through YouTube API. Also, we are fetching the followers count of an actor of the movie from the Twitter API by passing the actor name. After we got both the actor and followers count we are doing sentiment analysis on the YouTube comments data and saving the number of positive, negative and neutral sentiments for each movie trailer. We are normalizing this sentiment count and normalizing the followers count for each movie and saving the result. The final equation is being calculated for each movie and the value is between the 0 to 10. After we got the weighted value for all the 10 movies, we are running K-Means clustering algorithm and categorizing the output in HIT, NEUTRAL and FLOP. We are sending output to the ElasticSearch and visualizing the output on the Kibana dashboard.