SenticAPI testing for polarity classification and emotion recognition on Reddit comments regarding ChatGPT

• Main Author: Tan, Mazel Jia Ying
• Other Authors: Erik Cambria
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Computer and Information Science
• Online Access: https://hdl.handle.net/10356/174992
• Institution: Nanyang Technological University

ChatGPT has gained significant popularity across diverse domains, ranging from personal usage to business applications. Despite its widespread popularity, opinions regarding ChatGPT remain varied and at times contradictory. This disparity in perspectives highlights the complexity surrounding understanding ChatGPT's functionality and effectiveness and discerning users' sentiments and attitudes towards ChatGPT remains a challenge due to the diverse range of experiences and use cases. Therefore, it is imperative to delve deeper into understanding the multifaceted perceptions surrounding ChatGPT to ensure informed decisions are made regarding the usage of ChatGPT. This project makes use of SenticNet, a knowledge base that makes use of symbolic and subsymbolic AI to increase the accuracy of NLP. Previous work with SenticNet APIs has primarily concentrated on product review datasets. Therefore this study aims to expand the domain of the API into text data sourced from discussion forums like Reddit. By venturing into this new domain, the aim is to assess the API's performance and its effectiveness in analyzing unprocessed opinions and sentiments. Furthermore, previous studies on ChatGPT opinions have only focused on 3 broad text classification categories - Positive, Neutral, and Negative. However, such an approach overlooks the nuanced emotions that may exist within these categories. This project seeks to go beyond the limitations of this conventional method by employing a more comprehensive approach to classify comments based on the range of emotions outlined in the Hourglass of Emotions model. By doing so, it aims to capture the subtleties and intricacies of human sentiment expressed in discussions about ChatGPT on platforms like Reddit, providing a more informed understanding of user perceptions and attitudes. Overall, the API has shown promising results in the domain of Reddit discussions.