FIELD: physics.
SUBSTANCE: invention relates to means of determining a relevance parameter for a digital content item and a system for realizing said method. Digital content element originates from the content channel associated with the recommendation system. Pool of users associated with the content channel is identified, wherein the user from the user pool is associated with the content channel. User pool is used to analyse and predict the relevance parameter. Relevancy parameter is further used to predict a relevance value of the digital content item for a user outside the user pool based on user interactions of the first user.
EFFECT: technical result is higher accuracy of issuing recommendations.
20 cl, 5 dwg
