The emergence of Generative Artificial Intelligence has spurred widespread discussion and debate across various platforms, notably X (Twitter), where diverse opinions converge in a digital space. This research explores public sentiment toward Generative AI by analyzing #GenerativeAI tweets over a designated period using a multifaceted sentiment analysis approach. The study begins with VADER for preliminary sentiment evaluation and integrates traditional machine learning models (Naive Bayes, Logistic Regression, SVM), BERT-based models (Twitter-roBERTa), and text analysis techniques including TF-IDF, word clouds, and n-grams. This research contributes to the broader field of sentiment analysis by evaluating the effectiveness of multiple analytical approaches in the context of social media discourse on emerging technologies. The findings offer insights for academia, industry, and policymakers by highlighting public sentiments and concerns related to Generative AI, informing decision-making processes and guiding future research. The methodology assigns sentiment scores, classifies sentiments, visualizes key lexical components, and identifies prevalent word combinations, providing a holistic view of public opinion on Generative AI.
Sarbpreet Ghotra (Fri,) studied this question.