Concept of an Agent-Based AI Approach for the Automated Evaluation and Clustering of Free-Text Responses in the Context of UX Data

The growing amount of user experience (UX) data, particularly free-text responses, makes it highly challenging to derive meaningful information. Conventional approaches to analysing these responses, such as manual coding or simple keyword clustering, are time-consuming and, in most cases, not scalable. In response to this, this paper proposes using an artificial intelligence (AI) method based on a UX research agent to automatically evaluate and cluster the responses. Using large language models (LLMs), the study will develop a scalable and effective approach to clustering and automate the data analysis process, enhancing the quality and speed of analysis. The researcher summarises the available literature on applying agent-based AI and LLMs to UX assessment. They note that this method can accurately assess user sentiment, detect underlying themes in the assessment and provide better clustering results. This methodology can help UX researchers to better understand user feedback and make better design choices, as demonstrated in this paper. The paper also addresses some of the main challenges, such as the issues of trust and explainability, and the digital divide, which could impact the effective use of AI in UX research. The findings of this research emphasise the potential of AI to transform the UX data analysis process, providing a system that is more effective, scalable and informative for assessing user feedback in real time. The paper concludes with a discussion of the implications for future UX studies and the necessity for continuous progress in AI technologies to overcome current obstacles. Keywords: Agent-based AI, Clustering, UX Data, Free-Text Responses, Large Language Models, Automated Evaluation, Sentiment Analysis, AI-driven Clustering.