Hybrid Multi-Criteria Decision-Making and Neural Network Modeling for Molecular Evaluation of Sulfur Derivatives in Drug Design

Machine learning has greatly improved the understanding of drug structures and their physiochemical properties in research. The use of advanced algorithms and models in machine learning has helped scientists predict how molecules behave, invent the best ways to make chemicals and uncover novel medicines more rapidly. These developments have made it possible to understand biological systems and chemical actions better, increasing the quality and speed of drug development. To predict the properties of various drugs containing sulfur ([Formula: see text]), this study used artificial neural networks as well as random forests and adaptive boosting techniques. By using these different machine learning methods improved how accurately, our models could predict the results and gave valuable clues about the physiochemical relationships. The framework applies multiple decision strategies including TOPSIS to organize drug recommendations by therapeutic efficacy, safety, and cost, matching 95% of the preferences made by real-world Sulfur based drugs. With the help of the TOPSIS method, we confront the uncertainties and different criteria, drug effectiveness, adverse effects, patient-relevant factors, and economic factors, found in sulfur medicines.