Computational Sentiment Analysis of Top Movies
Advisor: Deepak Kumar, Computer Science & Linguistics
Sentiment analysis refers to the application of natural language processing, computational linguistics, and text analytics to automatically extract opinions, emotions and sentiments in a text. It aims to determine the attitude of a speaker or a writer with respect to some topic or the overall contextual polarity of a document. The attitude may be their judgment or evaluation, affective state (which is the emotional state of the author when writing), or the intended emotional communication.
With the rise of social media, such as blogs and social networks, sentiment analysis has become increasingly important and relevant in the 21st century. Through sentiment analysis, we can track attitudes and feelings on the web through posts on Facebook, Twitter, Instagram etc. Businesses can track the reviews, ratings, recommendations and other forms of online opinion to follow new product perception, track negative feedback, identify new opportunities and manage their reputation. It is currently an emerging area in natural language processing and computational linguistics and has received a lot of positive media attention as well due to its having a variety of practical applications. Using sentiment analysis on a social media platform like Twitter to predict stock market trends1 is one such example.
For my summer research I will be doing sentiment analysis of tweets associated with the top five movies of 2012. I will be using Pattern2, which is a web mining module for the Python programming language. Pattern enables retrieval of individual tweets from Twitter which can then be analyzed. The tweets will be rated based on a comprehensive rating scale. These ratings will contribute to the overall aggregate score of each movie. These ratings will be compared against the ratings available on Metacritic3, a movie review website.
1 Rakshit, Audrika. "Twitter Chatter and Stock Market Trends." Web log post. Masters of Media. University of Amsterdam, 15 Oct. 2012. Web. 4 June 2013. <http://mastersofmedia.hum.uva.nl/2012/10/15/twitter-chatter-and-stock-market-trends/>.
2 Smedt, Tom De. "Pattern." Computational Linguistics & Psycholinguistics Research Center. University of Antwerp, 26 Nov. 2011. Web. 4 June 2013. <http://www.clips.ua.ac.be/pattern>.
3 Metacritic. CBS Interactive, Jan. 2001. Web. 04 June 2013. <http://www.metacritic.com/>.