FIELD: banking security.
SUBSTANCE: invention relates to solutions for secure user interaction with remote services. A method for detecting fraudulent activity, when a user interacts with remote services, contains following stages: the initial classification of a remote service site is performed, while creating a semantic model of the remote service site; signs of user activity are identified, when the user interacts with the remote service site, using a script executed in the browser on the user’s computing device; a resource identifier is determined based on identified data using the semantic model of the remote service site; identified signs of user activity are analyzed, comparing identified signs of user activity and signs of early user activity with known scenarios of fraudulent activity; the fraudulent activity of the user is detected, if signs of current user activity and signs of early user activity correspond to the known scenario of fraudulent activity.
EFFECT: ensuring the claimed purpose.
4 cl, 3 dwg, 3 tbl
