Optimization of multi-element geochemical anomaly recognition in the Takht-e Soleyman area of northwestern Iran using swarm-intelligence support vector machine

A groundbreaking approach using the grasshopper optimization algorithm combined with support vector machine (SVM) has achieved over 95% accuracy in detecting multi-element geochemical anomalies at Takht-e Soleyman, a UNESCO World Heritage Site in northwestern Iran. This method eliminates the need for trial-and-error hyperparameter tuning, significantly reducing training time while maintaining high precision. The optimized SVM model, utilizing polynomial and radial basis kernel functions, not only advances mineral exploration under barren cover but also sets a precedent for applying swarm-intelligence techniques in geoscientific applications. This innovation could revolutionize the identification of undiscovered mineral deposits, addressing a critical challenge in industrial progress.

Sabbaghi, H., Tabatabaei, S. H., & Fathianpour, N. (2025). Optimization of multi-element geochemical anomaly recognition in the Takht-e Soleyman area of northwestern Iran using swarm-intelligence support vector machine. Frontiers in Earth Science, 13. https://doi.org/10.3389/feart.2025.1352912