Comparison of the performances of artificial intelligence bots using continuous intuitionistic fuzzy evaluation based on distance from average solution method

Alkan N., AYDIN U., Menekşe A., Kahraman C.

Engineering Applications of Artificial Intelligence, cilt.161, 2025 (SCI-Expanded) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 161
  • Basım Tarihi: 2025
  • Doi Numarası: 10.1016/j.engappai.2025.112033
  • Dergi Adı: Engineering Applications of Artificial Intelligence
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Aerospace Database, Applied Science & Technology Source, Communication Abstracts, Compendex, Computer & Applied Sciences, INSPEC, Metadex, Civil Engineering Abstracts
  • Anahtar Kelimeler: Artificial intelligence, Chat generative pre-trained transformer, Chatbot, Continuous intuitionistic fuzzy sets, Evaluation based on distance from average solution, Multi-criteria decision-making
  • İstanbul Ticaret Üniversitesi Adresli: Evet

Özet

The rapid evolution of artificial intelligence (AI) has introduced novel opportunities and challenges in various fields. In this study, we present a pioneering approach known as Continuous Intuitionistic Fuzzy (CINFU) Evaluation based on Distance from Average Solution (EDAS), an innovative extension of the EDAS method tailored to Continuous Intuitionistic Fuzzy Sets. This methodology is designed to compare the performance of AI tools. The capabilities of AI bots have been examined through their success rates in various tasks and uncertainty levels in decision-making processes. The study aims to evaluate the effectiveness of different models in decision-making processes by analyzing the performances of AI bots such as Chat Generative Pre-trained Transformer (ChatGPT), Bard, and Claude based on both objective measurements and fuzzy evaluation criteria. The comparison focuses on key performance criteria such as Bots Triggered, User Engagement, Message Click-Through Rate, Chat Handoff, User Retention, Bounce Rate & Dwell Time, Leads Captured, and Customer Satisfaction Score. Ultimately, the validity and robustness of the approach have been tested with sensitivity analysis.