In this context,two different approaches of soil liquefaction evaluation using a soft computing technique based on the worldwide standard penetration test(SPT) databases have been studied.Gene expression programming(GEP) as a gray-box modeling approach is used to develop different deterministic models in order to evaluate the occurrence of soil liquefaction in terms of liquefaction field performance indicator(LI) and factor of safety(F_S) in logistic regression and classification concepts.The comparative plots illustrate that the classification concept-based models show a better performance than those based on logistic regression.In the probabilistic approach,a calibrated mapping function is developed in the context of Bayes’ theorem in order to capture the failure probabilities(P_L) in the absence of the knowledge of parameter uncertainty.Consistent results obtained from the proposed probabilistic models,compared to the most well-known models,indicate the robustness of the methodology used in this study.The probability models provide a simple,but also efficient decision-making tool in engineering design to quantitatively assess the liquefaction triggering thresholds. In this context, two different approaches of soil liquefaction evaluation using a soft computing technique based on the worldwide standard penetration test (SPT) databases have been studied. Gene expression programming (GEP) as a gray-box modeling approach is used to develop different deterministic models in order to evaluate the occurrence of soil liquefaction in terms of liquefaction field performance indicator (LI) and factor of safety (F_S) in logistic regression and classification concepts. The comparative plots illustrate that the classification concept-based models show a better performance than those based on logistic regression. In the probabilistic approach, a calibrated mapping function is developed in the context of Bayes’ theorem in order to capture the failure probabilities (P_L) in the absence of the knowledge of parameter uncertainty. Consistent results obtained from the proposed probabilistic models, compared to the most well-known models, indicate the robustness of the methodology used in this study.The probability models provide a simple, but also efficient decision-making tool in engineering design to quantitatively assess the liquefaction triggering thresholds.