The failure of Reinforced concrete (RC) interior beam-column joints is quite common in structures during earthquake. So, the determination of failure mechanism of beam-column joint is an important task in structural engineering. This study employs the support vector machine (SVM) and least square support vector machine (LSSVM) models to determine the type of failure at an interior beam-column joint. SVM and LSSVM are firmly based on the theory of statistical learning, using classification technique. In this study, geometric, loading and material parameters are considered as input variables of SVM and LSSVM models. Equations are developed to identify the type of initiation of failure in interior beam-column joint and also for redesign purpose. Paper also describes a comparative study between the developed SVM and LSSVM models. © 2011 CAFET-INNOVA TECHNICAL SOCIETY.