In recent literatures, several approaches were proposed for learning Bayesian networks. These methods combine prior domain knowledge and data to derive one or more Bayesian networks. The Bayesian networks can be used to infer causal relationships among variables of interest. This thesis examines the most important factors of financial crisis from the aspect of fundamental analysis, as well as using the Bayesian networks to explore the causal relationship between the causes (financial ratios) and the effect (financial failures). The empirical results indicate that the average accuracy of prediction is about 80%.