In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
will defend his proposal
Characterizing Movies Using Plot and Reviews
AbstractIn this thesis, we propose to create a natural language processing system to generate a set of tags for movies by analyzing the written plot synopses and user reviews. These tags will be able to represent a diverse set of attributes of movies and the possible experiences from watching a movie in a concise way, that will eventually be beneficial for general users to make better choices when it comes to picking a movie to watch. Selecting a good movie is hard. Our choices depend on a wide range of factors like our background, age, mood, personality, or the company we keep when we watch a movie. Traditional recommendation systems fail to provide users sufficient insights about the stories of movies to let them know why they would like a particular movie, given any mental state. We expect our system would be able to let users know helpful characteristics about the story of movies that can play an important role in picking the right movie. As part of this thesis, we will work on developing computational methods to characterize narrative texts like the plot synopsis of movies. We will also develop extracting story-related attributes from user reviews posted in movie recommendation websites. Finally, we plan to release the end system for general users to use to get insights about movies and pipeline of relevant tools for researchers and developers to facilitate works relevant to this problem area.
Date: Wednesday, May 1, 2019
Time: 12:00 - 1:00 PM
Place: PGH 550
Advisor: Dr. Thamar Solorio
Faculty, students, and the general public are invited.