Off-campus SJSU users: To download campus access theses, please use the following link to log into our proxy server with your SJSU library user name and PIN.

Publication Date

Fall 2012

Degree Type

Thesis - Campus Access Only

Degree Name

Master of Arts (MA)




Hahn Koo


analysis, hashtag, microblog, sentiment, sentiment analysis, Twitter

Subject Areas



Microblogging has become the new and informal venue for communication and a main source of news and information for millions of users who subscribe to and submit short messages on a range of topics. Sentiment analysis is an attempt to determine the attitude of the author of the document to the topic, or the overall polarity of the document. While sentiment analysis using computational linguistics methods has proven to be extremely useful in examining large texts, analyzing the sentiment of microblogging messages presents some unique challenges to previously established methods. Extensive research has been undertaken in order to establish optimal ways to determine polarity and sentiment of messages; however, most approaches have either completely ignored or placed extreme limitations on the use of hashtags. By allowing unrestricted participation of hashtags in sentiment analysis, the results of this study demonstrated that hashtags contributed significant value to overall sentiment analysis of microblogs and consequently improved the sentiment scores results.