MSc Thesis Proposal Announcement of Jalpa Deepak Patel: " Revenue Forecast Prediction using Graphs and Deep Learning "

 

Date: Thursday August 12, 2021 

Time:  11:00am – 12:30pm 

Meeting URL:  https://zoom.us/j/91949013791?from=addon 

Passcode:    If interested in attending this event, contact the Graduate Secretary at csgradinfo@uwindsor.ca

 

Predicting the revenue of a movie is different from traditional revenue forecasting problems. Even though movies are among the top sources of entertainment, many movies involving talented star cast, directors and production companies do not have substantial financial success. This study explores the different factors that can help in predicting the extent of success of a movie. The social connections among actors, directors, and producers enhance their creativity and impact the making of a movie. Hence, this study uses the social connections among actors, directors and production companies by representing them in a graph. The sentiment of a movie’s plot to predict the revenue of a movie has been introduced in this research to highlight the emotional connection of the audience with the story. Finally, a Deep Learning model is proposed to forecast the revenue of a movie. The performance of the proposed model is compared with other Machine Learning models to determine the efficiency of the proposed model. 

 

Keywords: Graphs, Embeddings, Sentiment Analysis, Deep Learning, Revenue Forecasting 

 

Internal Reader: Dr. Dima Alhadidi    

External Reader: Dr. Roberto Muscedere 

Advisor: Dr. Dan Wu