AI-Facilitated Selection of the Optimal Nondominated Solution for a Serious Gaming Information Fusion Module

Fuzzy optimization problems play a substantial role in Information Fusion (ISO4: Inf Fusion) and related Multi-Criteria Decision Making (MCDM) problems. These problems tend to be complex because the measures/objectives tend to conflict with each other. Researchers have utilized various metaheuristic approaches for addressing these challenges, such as the obtaining of the Optimal Shapley-Nondominated Solution (OSNS) to contend with the Fuzzy Number (FN) coefficient issue, among others. In this paper, a bepoke Robust Convex Relaxations (RCR)-centric Particle Swarm Optimization (PSO) metaheuristic approach is utilized along with a unique Lower Ambiguity, Higher Uncertainty (LAHU) and Higher Ambiguity, Lower Uncertainty (HALU) Module (LHM) for not only seeking the OSNS, but also an enhanced approximation of nonlinear Spherical FNs (SFNs) via an Optimal Corresponding Generalized Linear “f”-sided SFN form (OCGLfSFN)-based membership function (in accordance with the involved ambiguity and uncertainty). The LHM seeks to ascertain whether an Isomorphic Paradigm (IsoP) has occurred before within the historical data (e.g., the IsoP has already been analyzed, gameplayed, etc.) via a bespoke Isomorphic Comparator Similarity Measure (ICSM) (for FNs/SFNs/[Spherical Fuzzy Sets] SFSs) that leverages OSNS and OCGLfFN.