Towards a UX data-centric approach: providing tools to support software development

• Ano de defesa: 2025
• Autor(a) principal: Macedo, Maylon Pires
• Orientador(a): Zaina, Luciana Aparecida Martinez
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal de São Carlos; Câmpus São Carlos
• Programa de Pós-Graduação: Programa de Pós-Graduação em Ciência da Computação - PPGCC
• Palavras-chave em Inglês: UX data; User experience; Information visualization; UX data taxonomy
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO::METODOLOGIA E TECNICAS DA COMPUTACAO
• Link de acesso: https://hdl.handle.net/20.500.14289/22947

The frequently collected data regarding user interaction with software (i.e., UX data) holds potential to enhance User Experience (UX), which encompasses users' emotional, practical, and perceived value interactions with products. However, a challenge exists between the collection of UX data and its effective analysis. Software practitioners frequently encounter difficulties in selecting relevant data and applying appropriate techniques to gain insights for software improvement. The field of Information Visualization (InfoVis) offers approaches to assist users in analyzing emerging information from data. While initial efforts explored how InfoVis principles could be applied, this research revealed a more fundamental problem. A Grey Literature Review on UX data not only identified diverse formats, definitions, and suggested visual representations but, exposed a pervasive lack of formal UX data definitions in practice. Two initial exploratory studies further reinforced these challenges, demonstrating that practitioners struggled to derive insights from unstructured UX data without a clear conceptual foundation. Building upon these findings, this research strategically pivoted its focus to address the core issue for effective UX data analysis. The central artifact developed is a UX data taxonomy, designed to organize and clarify UX data understanding by categorizing objectives, purposes, and data characteristics. This taxonomy itself represents a theoretical contribution, providing a structured model for the UX data domain. To ensure its practical utility and accessibility, this taxonomy was operationalized through two distinct tools: an interactive web-based system for exploration and a Generative AI (GenAI) RAG (Retrieval-Augmented Generation) chat tool for natural language interaction. The utility and acceptance of this taxonomy and its related tools were then empirically validated with software professionals. Finally, the research also explored emerging opportunities for UX data analysis, particularly concerning the role of GenAI, assessing how these advancements intersect with the need for structured data understanding, thus providing theoretical insights into the evolving landscape of UX data analysis. This thesis provides both practical tools to empower software professionals in leveraging UX data for enhancing interactive system design, and lays a theoretical foundation for understanding and working with UX data.