Integrating Multi-Label Classification and Generative AI for Scalable Analysis of User Feedback

The paper introduces a framework that integrates multi-label classification with generative AI to analyze user feedback more efficiently, particularly in the context of UX evaluation. It utilizes supervised machine learning to assign predefined topic labels to user comments and employs generative AI to summarize these insights, enhancing communication of findings. The study also finds that sentiment analysis is insufficient for gauging overall product satisfaction, emphasizing the need for explicit survey metrics. This approach offers practitioners a scalable method to process qualitative feedback, improving UX assessment and product development strategies.