Relevance Vector Machine for Evaluating Seismic Liquefaction Potential Using Shear Wave Velocity

In this paper, the potential of relevance vector machines (RVM) based classification approach has been used to assess the liquefaction potential from actual shear wave velocity data. RVM is based on a Bayesian formulation of a linear model with an appropriate prior that results in a sparse representation. As a consequence, it can generalize well and provide inferences at low computational cost. It also gives probabilistic output through Bayesian inference. Input parameters of RVM model are effective vertical stress (σ'_vo), peak acceleration at the ground surface (a_max), earthquake magnitude (M), shear wave velocity (V_s), and soil type. In this study, Gaussian functions are used as kernel for RVM model. The results show that the RVM model predicts liquefaction with accuracy of 98.21%. The developed RVM model provides a viable tool to earthquake engineers in assessing the sites susceptible to liquefaction.