A newly developed multi-criteria group decision-making framework provides a reliable and flexible approach for evaluating and selecting machine learning techniques in uncertain environments, according to a study published in Discover Applied Sciences. Selecting suitable machine learning techniques requires an intelligent decision-support framework capable of effectively handling uncertainty, ambiguity, and expert preferences across multiple evaluation criteria. To address these challenges, researchers Muhammad Ismail, Tahir Zaman, Zabihullah Movaheedi, and Asghar Khan developed a novel decision model operating under circular bipolar fuzzy information.
Multi-criteria group decision-making represents an established approach for solving complex decision-making problems that involve multiple competing criteria alongside uncertain information. In the researchers’ model, uncertainty and ambiguity are handled through the circular bipolar fuzzy set. The circular bipolar fuzzy set incorporates positive membership, negative membership, and the radius between these degrees. According to the study, combining these three components provides a flexible mathematical representation of uncertain and bipolar information during the evaluation process.
To process the circular bipolar fuzzy information, the authors defined novel Schweizer–Sklar operational laws. These operational laws were formulated specifically using the Schweizer–Sklar t-norm and t-conorm to govern computations with circular bipolar fuzzy data. Using these operational principles, the team developed an aggregation tool designated as the Circular Bipolar fuzzy Schweizer–Sklar Weighted Averaging operator. This operator aggregates individual expert evaluations into collective assessments while preserving the bipolar and uncertain structure of the input numbers.
For ranking alternatives and determining the most appropriate machine learning technique, the framework implements an EDAS-based multi-criteria group decision-making approach using circular bipolar fuzzy numbers. This decision-making procedure evaluates candidate machine learning techniques by accounting for expert preferences and ambiguity across multiple criteria. The circular bipolar fuzzy numbers allow the decision framework to maintain both positive and negative membership degrees along with their associated radius throughout the ranking calculations.
To validate the performance of their proposed decision framework, the researchers conducted comparative and sensitivity analyses. These testing procedures evaluated both the effectiveness and the robustness of the decision-making model across uncertain and bipolar environments. The authors report that the findings demonstrate the model delivers a reliable and flexible methodology for selecting suitable machine learning techniques when facing complex, multi-criteria decision scenarios involving ambiguity and conflicting expert input.
Subject of Research: Machine learning technique selection using circular bipolar fuzzy decision-making
Article Title: An intelligent decision-making model under circular bipolar fuzzy information with application to machine learning techniques selection
Article References: Ismail, M., Zaman, T., Movaheedi, Z., & Khan, A. (2026). An intelligent decision-making model under circular bipolar fuzzy information with application to machine learning techniques selection. Discover Applied Sciences. https://doi.org/10.1007/s42452-026-09688-6
Image Credits: AI Generated
DOI: 10.1007/s42452-026-09688-6
Keywords: multi-criteria group decision-making, circular bipolar fuzzy information, circular bipolar fuzzy set, machine learning technique selection, Schweizer–Sklar operational laws, Schweizer–Sklar t-norm, Schweizer–Sklar t-conorm, C-BFSSWA operator, EDAS-based MCGDM, circular bipolar fuzzy numbers, positive membership, negative membership, bipolar information, expert evaluations, sensitivity analysis, comparative analysis, decision-support framework, uncertainty modeling
News Source: Teresa Odom. (November 18, 2026). Circular Bipolar Fuzzy Decision Model Evaluates Machine Learning Techniques Under Uncertainty. Scienmag.



