Exploring ChatGPT’s Role in Education: An NLP-Based Sentiment and Thematic Study Using Twitter Data

Mohd Danish, Mohammad Amjad, Kanagat Dyussekeyev, Аinur Zhumadillayeva

2025

Full text Open access: Gold Metadata (CSL-JSON)
DOI 10.1016/j.procs.2025.04.049
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

The emergence of AI-driven technologies such as ChatGPT in education has profoundly influenced students’ learning experiences, especially in India. This study examines the sentiment of ChatGPT’s application in Indian education by evaluating 25,000 tweets gathered from August 2023 to June 2024 utilizing Natural Language Processing (NLP) techniques. The data underwent noise reduction, followed by sentiment analysis utilizing both lexicon-based (TextBlob) and machine learning models (BERT and RoBERTa) for a comprehensive assessment. Thematic analysis utilizing Latent Dirichlet Allocation (LDA) was employed to discern major topics including educational improvement, ethical issues, and the future of AI integration in education. Findings indicate that 45% of the tweets exhibited positive sentiment, emphasizing advantages in tailored learning and accessibility, whereas 30% reflected negative sentiment, frequently addressing issues related to academic integrity and prejudice in AI models. A comparative investigation of sentiment analysis models demonstrated that RoBERTa surpassed others with a 92% accuracy rate, confirming its exceptional contextual comprehension. Ethical issues, including data protection, possible misuse, and excessive dependence on AI, were substantial, leading to suggestions for the ethical implementation of ChatGPT in educational contexts. This study enhances the conversation on AI in education by elucidating the perceptions of students and educators in India about ChatGPT, while underscoring the necessity for ethical rules and policies.