The construction industry's pursuit of sustainability has intensified the development of green concrete incorporating multiple waste materials as partial cement replacements. However, the complex, non-linear relationships between mixture proportions and performance characteristics of these hybrid systems present significant challenges for traditional empirical modeling. This comprehensive review presents a data-driven framework integrating Weight of Evidence (WoE) and Artificial Neural Networks (ANN) to predict the mechanical and durability properties of hybrid green concrete containing combinations of industrial and agricultural wastes. The paper systematically analyzes how WoE methodology identifies and quantifies the influence of key mixture parameters including replacement types (Rice Husk Ash (RHA), Sugarcane Bagasse Ash (SCBA), Fly Ash, Waste Glass Powder (WGP)), replacement levels, water-binder ratios, and curing conditions on concrete performance. The review demonstrates how these prioritized factors then serve as optimized inputs for ANN models, creating highly accurate predictive systems for compressive strength, tensile strength, permeability, and chemical resistance. By synthesizing findings from extensive experimental studies on binary, ternary, and quaternary cement replacement systems, this review establishes that the WoE-ANN integration achieves prediction accuracies of 92-97% for mechanical properties and 85-90% for durability indicators, significantly outperforming conventional regression models. The framework provides researchers and practitioners with a powerful methodology for optimizing complex hybrid mixtures, accelerating the development of sustainable concrete formulations while ensuring reliable performance. This approach represents a paradigm shift from trial-and-error experimentation to intelligent, data-driven design of next-generation green construction materials.