Object recommendation method, apparatus, medium, and equipment

Abstract

A subject recommendation method includes calculating a first score value and a second score value corresponding to each object in object set, calculating a comprehensive score value corresponding to each object according to the first score value and the second score value corresponding to each object, and determining target object to be recommended to target user according to the comprehensive score value corresponding to each object. In this method, the algorithm based on user cooperation and the core technology based on popular object recommendation mechanism are used, and are proposed to measure the degree that the object accords with the interest tendency of the target user according to the scores of the two dimensions. As the comprehensive scores of the object, it can be ensured thatthe determined target object is the hot object with the highest degree of coincidence with the interest of the target user and can arouse the interest of the user, so as to ensure the accuracy of therecommendation results. An object recommending apparatus, equipment, and a computer storage medium are also disclosed.

Description

technical field [0001] The present application relates to the technical field of the Internet, in particular to an object recommendation method, device, medium and equipment based on a collaborative filtering algorithm. Background technique [0002] With the rapid development of Internet technology, a variety of network platforms have entered people's daily life and work, bringing many conveniences to people's work and life; for example, e-commerce platforms, news platforms, video Platforms, service recommendation platforms, etc. In order to provide platform users with more personalized and accurate services, these network platforms often need to use recommendation systems to achieve accurate personalized recommendations. [0003] The recommendation system that is widely used at this stage is based on collaborative filtering algorithm. Its working principle is collaborative filtering based on user collaboration. First, it finds similar users of the target user according to t...

AI Technical Summary

Problems solved by technology

[0004] Since the current recommendation system is only based on the dimension of user collaboration, it selects the target object to be recommended for the target user from the objects that the target user is interested in, and does not consider the factors that are recommended for the target user from the dimension of popular objects. Therefore, In practical applications, the recommendation results of the recommendation system cannot recommend novel objects that users are interested in, and do not meet the business needs of the network platform, resulting in the loss of network platform users