Exploring User Sentiments on Artificial Intelligence for Education: A Topic Modeling Analysis of Google Play Reviews

Abstract

The rapid advancement and integration of Artificial Intelligence (AI) tools, such as ChatGPT, are significantly reshaping the education sector. As a widely adopted AI application, ChatGPT offers substantial potential for enhancing digital education. Understanding user perceptions is crucial for optimizing its role in educational settings. Given the extensive volume of user feedback on Google Play, automated analysis is essential to extract meaningful insights. This study employs an aspect-based sentiment analysis approach, utilizing Latent Dirichlet Allocation (LDA) for topic modeling and Naïve Bayes for sentiment classification. The analysis identified five key aspects influencing user sentiment: Usability, Response, Feature, Functionality, and Satisfaction, achieving a coherence score of 0.524316. While positive sentiment generally dominates across these aspects, the Response aspect received the highest negative feedback, particularly concerning accuracy and response time. Additionally, data imbalance posed challenges in detecting negative sentiment. These findings offer valuable insights into the adoption of AI in education, highlighting ChatGPT's strengths and identifying areas for improvement to foster more effective and engaging learning experiences.

Ähnliche Arbeiten