Quantifying Ozempic’s Impact: Sentiment-Based Drug Evaluation with BERT and Mistral Models

Session Number

CMPS 28

Advisor(s)

Dr. Ashiqur R. KhudaBukhsh, Rochester Institute of Technology

Discipline

Computer Science

Start Date

17-4-2025 11:40 AM

End Date

17-4-2025 10:55 AM

Abstract

Semaglutide, sold under the name Ozempic, is a medication that aids in blood sugar management as well as weight loss in individuals diagnosed with type 2 diabetes. Ozempic is frequently advertised as an effective weight loss drug due to the uncontrolled popularity stemming around it from social media.

The purpose of this study is to understand the public opinion and real-life testimonials regarding the use of Ozempic via Youtube comments. I trained BERT and Mistral models for sentiment classification and thematic pattern recognition using a labelled dataset of generated content. Three major perspectives were identified to focus on: users of Ozempic, the families of users of Ozempic, and the general public. This study utilized natural processing language (NLP) to analyze sentiment patterns, issues repeated over time, and the general perception of the effectiveness of the drug of different people. The results provide information on the qualitative effects experienced by users of Ozempic outside the clinical setting and the views that shape the perception of Ozempic in public.

Share

COinS
 
Apr 17th, 11:40 AM Apr 17th, 10:55 AM

Quantifying Ozempic’s Impact: Sentiment-Based Drug Evaluation with BERT and Mistral Models

Semaglutide, sold under the name Ozempic, is a medication that aids in blood sugar management as well as weight loss in individuals diagnosed with type 2 diabetes. Ozempic is frequently advertised as an effective weight loss drug due to the uncontrolled popularity stemming around it from social media.

The purpose of this study is to understand the public opinion and real-life testimonials regarding the use of Ozempic via Youtube comments. I trained BERT and Mistral models for sentiment classification and thematic pattern recognition using a labelled dataset of generated content. Three major perspectives were identified to focus on: users of Ozempic, the families of users of Ozempic, and the general public. This study utilized natural processing language (NLP) to analyze sentiment patterns, issues repeated over time, and the general perception of the effectiveness of the drug of different people. The results provide information on the qualitative effects experienced by users of Ozempic outside the clinical setting and the views that shape the perception of Ozempic in public.