Abstract: Disclosed are system and method for identifying suspicious user behavior during a user's interaction with various banking services. One exemplary method comprises: receiving information relating to user's interaction with two or more banking services from at least two computing devices used by a user for interacting through a user account with each banking service; receiving an identifier of each computing device; determining a model of user behavior based at least on received information and identifers; calculating a probability of fraud based at least on the model of user behavior; determining and forming patterns of suspicious user behavior; and determining whether a current user activity in interacting with at least one banking services is suspicious based at least upon the patterns.