Akademik Veri Yönetim Sistemi
7th International Conference on Intelligent and Fuzzy Systems, INFUS 2025, İstanbul, Türkiye, 29 - 31 Temmuz 2025, cilt.1528 LNNS, ss.698-706, (Tam Metin Bildiri)
Cyber-Physical Systems (CPS) require intelligent agents capable of reasoning in uncertainty and adapting to dynamic environments. Fuzzy-BDI agents, which combine fuzzy logic with the Belief-Desire-Intention (BDI) paradigm, offer a robust solution but introduce additional complexity due to their hybrid nature. One of the most challenging aspects of their development is the manual creation of fuzzy rules from raw data collected by sensors, which is time-consuming and error-prone. This exploratory research proposes automating fuzzy rule extraction for fuzzy-BDI agents, streamlining the integration of fuzzy reasoning into BDI agents. The proposed approach involves extracting knowledge from the sampled data using machine learning algorithms to rigorously derive fuzzy rules, which are then embedded in the agent’s belief base. By automating this process, the framework reduces human effort, decreases time consumption, and improves the usability of fuzzy-BDI in CPS applications. The study is evaluated by comparing the manual definition of rules and the automated extraction of rules.