The current global financial crisis has demonstrated the importance of understanding sources of domestic and global vulnerabilities that could lead to a systemic event. In this paper, we use the Selforganizing map (SOM) for visualizing systemic risks and predicting systemic events. The SOM model allows disentangling the individual sources of vulnerabilities as well as performs equally well as a logit model in predicting out-of-sample the systemic financial crisis of 2007--2009. Further, we introduce the concept of the financial stability cycle and show how the SOM model can be used for monitoring it on a two-dimensional plane.