Exploring the Evolving Perception of Sports Doping through Text Analytics: Figuring out Public Sentiments

Authors : Sukumaran C, Ahmet Yavuz Karafil, Marimuthu K, Sugumar Chokkalıngam

Pages : 1-11

Doi:10.5281/zenodo.10899883

View : 499 | Download : 232

Publication Date : 2024-06-26

Article Type : Research Paper

Abstract :This study employs text analytics to investigate the evolving public perception of sports doping, utilizing natural language processing techniques such as sentiment analysis and Latent Dirichlet Allocation (LDA) topic modeling. The data, comprising 64,725 English tweets collected from January 2018 to December 2022 using the snscrape Python library, underwent refinement through NLP algorithms, including the removal of stop words, punctuation, digits, and hyperlinks. Stemming and lemmatization techniques were applied for text uniformity and structure enhancement. The sentiment analysis revealed dynamic shifts in neutral, pessimistic, and optimistic sentiments across different years, indicating changing public attitudes toward sports doping. Visual representations through figures and tables enhance comprehension of the sentiment distribution trends. LDA topic modeling identified critical themes in the sports doping discourse, encompassing anti-doping regulations, financial impacts on sports, marijuana use, women\'s participation, dope screening procedures, ethical considerations, anti-doping organizations, specific doping instances, and consequences of doping bans. The research illuminates a nuanced understanding of the intricate issue by presenting these findings visually. The study highlights a notable shift in public attitudes towards sports doping. It underscores the dynamic nature of opinions, emphasizing the importance of continuous observation and comprehension of factors influencing these fluctuations. The LDA analysis provides valuable insights into various dimensions of the sports doping debate, contributing to a more comprehensive understanding of this complex and evolving issue.
Keywords : Sports doping, sentiment analysis, text analytics, topic modeling, Latent Dirichlet Allocation