The integration of Large Language Models like ChatGPT into machine learning workflows represents a transformative shift in how sentiment classification models are developed, making advanced artificial intelligence accessible to those without extensive programming expertise. Through structured prompting strategies, including Task-Actions-Guidelines (TAG) and Persona-Instructions-Context (PIC) frameworks, individuals with basic computational thinking can now navigate complex technical processes from data preprocessing to model evaluation. This democratized paradigm demonstrates comparable performance to traditional expert-developed solutions while dramatically reducing development time and resource requirements. Beyond technical performance, ChatGPT-guided development offers enhanced interpretability, comprehensive documentation, adaptability to changing requirements, and significant educational benefits. The resulting paradigm shift creates new opportunities across educational settings, enables interdisciplinary collaboration, accelerates implementation in industry contexts, and raises important ethical considerations around responsible AI development. By lowering technical barriers while maintaining output quality, this innovation expands participation in machine learning development to previously excluded groups, potentially unleashing diverse perspectives that will drive the next wave of innovation in artificial intelligence applications.
Keywords: ChatGPT, low-code development, machine learning democratization, natural language processing, sentiment classification
