FIELD: physics.
SUBSTANCE: invention relates to computer engineering. Technical result is achieved by receiving, by means of a server from a web server, a first request for a target message, wherein the first request includes a target context parameter and a target minimum price, execution by a machine learning algorithm (MLA), wherein the MLA is configured to generate, based on the target context parameter and the target reserve price, a first reliability parameter indicating the probability that one message from the first plurality of target messages arranged on the first target message server is a target message corresponding to at least the first condition and second condition, wherein the first condition corresponds to the target message context parameter corresponding to the target context parameter, and the second condition corresponds to the bid price exceeding the target minimum price, of the second reliability parameter indicating the probability that one message from the second plurality of target messages placed on the second target message server, is a target message corresponding to at least said first condition and second condition.
EFFECT: accelerating the process of delivering a target message for inclusion in a web resource with simultaneous increase in the accuracy of selecting a target message relevant to given two conditions, and reducing the load on computational power.
20 cl, 8 dwg
