The USF School of Information showcased its expertise at the 37th Annual International 51在线 Artificial Intelligence Research Society (FLAIRS) Conference. The poster titled: 鈥淗ow did an election fraud narrative spread online? Testing theories using machine learning and natural language processing鈥� received the Best Poster Award. Co-authored by USF Associate Professors Dr. Loni Hagen, Dr. Joshua Scacco, and Assistant Professor, Dr. Ly Dinh, along with undergraduate research assistants Diego Ford and Jahnae Edwards from the BSIS Data Science concentration, and Dr. Nic DePaula, Assistant Professor at SUNY Polytechnic Institute, this acclaimed poster stood out among the submissions. Additionally, USF submitted another notable poster titled 鈥淕overnment Health Communication During the COVID-19 Pandemic: A BERT Topic Modeling Approach,鈥� co-authored by Mr. Thomas Moore-Pizon, Dr. Hagen, and Dr. DePaula.
Mr. Ford presented the award-winning poster at the FLAIRS conference. Despite it being his first time attending and presenting at the event, his presentation attracted significant interest from attendees. Their research focused on examining how disinformation spreads through social media. Mr. Ford explained, 鈥淲e specifically looked at Twitter from the period of November 1-15, 2020.鈥� The team placed particular emphasis on tweets mentioning dominion voting and investigated the impact for disinformation dissemination.
Ms. Edwards and Mr. Ford reflected on their surprise at the extent of the 鈥淭rump Effect,鈥� with Ms. Edwards noting, 鈥淲e noticed that his tweets lead to changes in the conversation surrounding the topic. Even the other most influential accounts were Republican accounts that were known for being large Trump supporters, some with past interactions with Trump on Twitter.鈥� Mr. Ford added, 鈥淚t is not surprising to see that Trump has a lot of influence in spreading misinformation in the 2020 election but the extent to which his influence made an impact was surprising. In our research, we called it the 鈥楾rump Effect鈥�. Essentially, we observed that small conversations could blow up into large national topics when Trump endorsed them, and they would then feed into smaller groups of conservative influencers across Twitter.鈥�
Both students shared their positive experience working as co-authors. Ms. Edwards reflected on their collaboration: 鈥淚t was a pleasure collaborating with my co-authors. We all got along very well, and our weekly meetings were a great help in efficiently completing this poster.鈥� Ms. Edwards commended Dr. Hagen for her support, commenting: 鈥淒r. Hagen is an incredible teacher who seamlessly combines getting insightful results in her research with teaching us how to further our coding skills. She was also always willing to hear our ideas and passions for possible additions to the poster. Diego and I worked well together, always willing to help each other when stuck on various tasks.鈥� Mr. Ford echoed the same sentiments, sharing: 鈥淚 loved working with my co-authors. They are very knowledgeable, and I have learned so much over the past year from them. Even though I am an undergraduate student, they still allowed me to contribute to the research, which is such a special opportunity.鈥�
Mr. Ford is a senior undergraduate student at USF, currently enrolled in the BSIS Data Science and Analytics program. Ms. Edwards graduated from USF鈥檚 undergraduate program with double majors in Information Science, concentrating in Data Science and Analytics, and Quantitative Economics and Econometrics. She will begin her master鈥檚 this fall in Analytics with a Computational Data Analytics track at Georgia Tech.