METHOD AND SYSTEM FOR IDENTIFYING CLUSTERS OF AFFILIATED WEBSITES Russian patent published in 2021 - IPC G06F16/00 G06F21/00 

FIELD: information technology.

SUBSTANCE: invention relates to a method and a system for determining affiliation of a web resource to clusters of affiliated web resources. In the method, at the training step, scanning communication networks for detecting a plurality of web resources; searching for information associated with each of plurality of web resources; generating corresponding patterns of each of plurality of web resources; grouping plurality of web resources into a plurality of clusters, wherein the grouping is based on affinity of plurality of web resource patterns; storing plurality of web resource clusters in memory; at working step obtaining URL link to previously unexplored web resource; searching for information on this web resource; generating a new web property data pattern; analyzing the affinity of the new pattern with patterns associated with the plurality of web resource clusters; calculating coefficient of affinity of pattern of given web resource with each of plurality of clusters; in response to the affinity coefficient exceeding the pattern with one of the plurality of clusters of the predetermined threshold value, associating said web resource with one specific cluster; updating a plurality of clusters stored in a database based on affinity of the web resource.

EFFECT: technical result consists in determining web property membership.

10 cl, 8 dwg