Using Twitter Sentiment Analysis to Analyze Self-Sentiment of the POTUS
Document Type
Conference Proceeding
Publication Date
2018
Publication Title
2018 Proceedings of the Conference on Information Systems Applied Research
ISSN
2167-1508
Abstract
This article provides a discussion of different tools that can be used to perform research on Twitter. The study then utilizes sentiment analysis to demonstrate self-sentiment of the POTUS and compare it with popular news sources. A comparison in 2 time periods and 4 months apart was made to determine if there is a change in the self-sentiment of the POTUS versus common news sources. To perform sentiment analysis, we utilize Python and the Vader and Pandas libraries, and statistical analysis was performed for each of the datasets. The first round of tests were based on a November dataset and revealed the means between the public, FOX, CNN and the President were not equal and that the POTUS had a higher self-sentiment than the sentiment of the news sources. The second round of tests were based on the following April data collection and revealed that the self-sentiment of the POTUS was not significantly different from the public or FOX. However, it was significantly different from CNN indicating a possible clash between the POTUS and CNN news.
Recommended Citation
White, Gwen, Hayden Wimmer, Carl Rebman, Chrisitan Sunday Nwankwo.
2018.
"Using Twitter Sentiment Analysis to Analyze Self-Sentiment of the POTUS."
2018 Proceedings of the Conference on Information Systems Applied Research: 1-11 Norfolk, Virginia: ISCAP, Information Systems and Computing Academic Professionals.
source: http://proc.conisar.org/2018/pdf/4812.pdf
https://digitalcommons.georgiasouthern.edu/information-tech-facpubs/109
Copyright
©2018 ISCAP (Information Systems & Computing Academic Professionals)
