Utilization of Statistical Learning Algorithms for Prediction of Elastic Modulus of Jointed Rock Mass

Pijush Samui

Abstract

This study uses two statistical learning algorithms for the prediction of elastic modulus (Ej) of jointed rockmass. The first algorithm uses support vector machine (SVM) that is firmly based on the theory of statisticallearning and uses regression technique by introducing -insensitive loss functionhas been adopted. Thesecond algorithm uses relevance vector machine (RVM). It is based on a Bayesian formulation of a linearmodel with an appropriate prior that results in a sparse representation. The RVM model gives variance ofpredicted data. The inputs of models are joint frequency (Jn), joint inclination parameter (n), joint roughnessparameter (r), confining pressure (3) and elastic modulus (Ei) of intact rock. Equations have been developedfor the determination of Ej of jointed rock mass based on the SVM and RVM models. The results of SVMand RVM models are compared with a widely used artificial neural network (ANN) model. This studyshows that the developed SVM and RVM models can be used for the prediction of Ej of jointed rock mass.Keywords: elastic modulus, jointed rock, support vector machine, relevance vector machine, artificial neuralnetwork