Intuitionistic complex Fuzzy sets in decision support systems: A choquet operated data mining-ANN approach

This work introduces a novel set, the Intuitionistic Complex Fuzzy Set (ICFS), that expands traditional intuitionistic fuzzy sets into a complex-valued domain and captures interacting attributes more effectively in the decision support framework based on ICFS. A new aggregation operator called the Intuitionistic Complex Fuzzy Einstein Correlated Geometric (ICFECG) operator and a new score and accuracy function for the ICFS are proposed and to ensure theoretical robustness, rigorous proofs are provided for multiple theorems associated with the newly developed ICFECG operator, the score and the accuracy functions. This operator effectively combines expert opinions while preserving both the amplitude and phase components of complex uncertainty, thereby ensuring that the aggregated information accurately reflects the full structure of the intuitionistic complex fuzzy evaluations. To improve efficiency in solving MAGDM problems, a data mining–based dimensionality reduction strategy that helps identify and remove redundant or weakly influential attributes is introduced. Artificial Neural Network (ANN) techniques are also incorporated to enhance the learning ability and optimization of the decision-support process. A new defuzzification function is proposed to integrate all the ICFS components, yielding a crisp value for enhancing the data mining and ANN computations. The final hybrid model combines ICFS theory, the ICFECG operator, data mining, and ANN optimization which effectively handles high-dimensional, correlated, and uncertain information arising in the decision making environment. A numerical case study shows that our methodology reduces the computational load, removes insignificant alternatives, and significantly improves decision accuracy, stability, and reliability.

Keywords:

Intuitionistic Complex Fuzzy set , MAGDM , Choquet integral operator , Data Mining , ANN