Date of Completion

2025

Document Type

Honors College Thesis

Department

Computer Science

Thesis Type

Honors College

First Advisor

Juniper Lovato

Second Advisor

Abigail Crocker

Keywords

MVP Sentiment Analysis Basketball Twitter

Abstract

The rapid growth of social media platforms like Twitter provides data scientists with unprecedented access to opinions on various subjects. In predictive modeling, these stores of diverse opinions could enhance traditional formulas or offer an alternative perspective to shed light on the potential for sentiment analysis in the prediction space. This paper discusses an effort to quantify the importance of player narrative in the NBA MVP race, a feature often absent in many past prediction models, by performing sentiment analysis on Twitter data. Using Python tools and logistic regression modeling, we collected and analyzed tweets and advanced statistics spanning five NBA seasons. The results are presented graphically to show the feature importance of advanced and sentiment statistics. Following the results, we discuss the power of sentiment analysis and its potential in the prediction space moving forward.

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.

Available for download on Friday, May 08, 2026

Share

COinS