Date Thesis Awarded


Access Type

Honors Thesis -- Access Restricted On-Campus Only

Degree Name

Bachelors of Science (BS)


Computer Science


James Deverick

Committee Members

James Deverick

Daniel Parker

Robert Michael Lewis


Microblogging has become one of the most useful tools for sharing everyday life events and news, especially popular sporting events, and for expressing opinions about those events. The English Premier League (EPL), the most popular professional soccer league in the world, is talked about on Twitter every day, and the 2015/16 season, whose title underdogs Leicester City managed to win, was one for the history books to remember. As Twitter posts are short and constantly being generated, they are a great source for providing public sentiment towards events that occurred throughout the 2015/16 EPL season. In this project, we examine the effectiveness of machine learning and text sentiment analysis on classifying the sentiment of tweets about Leicester City. We accomplish this by collecting tweets containing the words “Leicester City” using the python library GetOldTweets3; manually labelling those tweets as positive, negative, or neutral; and training an SVM classifier to classify tweets about Leicester City from the 2015/16 season. Our model achieved an F1-score of 0.76. We use the sentiments returned from the classifier to find correlations between real-life events and sentiment changes throughout the whole season and during individual games. From our analysis, we discovered an increase in tweets about Leicester City but a sentiment change from positive to negative as the season progressed. We also observed a wide range of changes in sentiment during a single match involving Leicester City due to real-life events as well as other factors which we discuss in detail.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 4.0 License.

On-Campus Access Only