METHOD FOR DETECTING PHISHING SITES AND SYSTEM THAT IMPLEMENTS IT Russian patent published in 2024 - IPC G06F21/50 

FIELD: information technologies.

SUBSTANCE: increased accuracy of detection of phishing websites is achieved through the stages of: forming a tree structure in the form of a DOM tree for each page; converting the generated DOM tree into a string consisting of DOM tree elements using a template; creating a hash for each generated string; search for the hash in the database; analysing the page, if no hash is found in the database, using a classifier including a machine learning model, and transmitting the information about the received page to the input of the machine learning model. With this purpose, first: train the specified classifier based on the generated and optimized data sets, verify the trained classifier on a test subset and transfer the trained classifier to work if the trained classifier does not exceed a given threshold when making incorrect decisions, or retrain the trained classifier if the trained classifier made the number of incorrect decisions exceeding the specified threshold; determine a site as phishing if the site page is recognized as phishing by the output of the machine learning model.

EFFECT: increased accuracy of detection of phishing websites.

10 cl, 9 dwg