본문 바로가기
  • Home

Fuzzy AHP-based Weight Analysis of Selection Attributes for ChatGPT and Gemini

  • Journal of The Korea Society of Computer and Information
  • Abbr : JKSCI
  • 2026, 31(10), pp.193~206
  • Publisher : The Korean Society Of Computer And Information
  • Research Area : Engineering > Computer Science
  • Received : August 21, 2026
  • Accepted : October 6, 2026
  • Published : October 30, 2026

Yujin Kim 1,  Hyung-Seok Lee 1

1충북대학교

Accredited

ABSTRACT

Extending prior research on the selection attributes of generative AI services, this study develops a two-level hierarchical model comprising four upper-level dimensions-functional attributes, experiential attributes, adaptive attributes, and brand image-and eight lower-level criteria, namely information accuracy, explainability, enjoyment, anthropomorphism, personalization, creativity, corporate reputation, and corporate innovativeness, to better understand the complexity of users' choice environments arising from the increasing diversity of generative AI services. To account for the imprecision inherent in human judgment, Fuzzy AHP was employed to derive the priority weights, and a conventional AHP analysis on the same data yielded nearly identical weights, thereby confirming the robustness of the Fuzzy AHP results. The analysis confirms that information accuracy emerges as the highest-priority attribute of generative AI services, consistent with prior findings. An evaluation of the two alternatives, ChatGPT and Gemini, showed that Gemini held a slight advantage over ChatGPT in terms of overall score. These findings provide empirical guidelines for service differentiation and strategic resource allocation in the design of future AI services.

Journal Copyright Policy

No CCL information provided

Citation status

* References for papers published after 2025 are currently being built.