Object recommendation method and device, electronic equipment and storage medium

Abstract

The invention discloses an object recommendation method. The method comprises the steps of obtaining current search content of a user; determining a to-be-recommended object set based on historical operation data of the user; determining the user association degree of the objects in the object set based on the search content and the feature information of the objects, wherein the feature information at least comprises purchase frequency; determining a penalty factor of the object according to the popularity of the object, wherein the popularity is negatively correlated with the penalty factor; determining a matching coefficient of the object based on the penalty factor and the user association degree, wherein the matching coefficient is used for representing the matching degree of the object and the user demand; and based on the matching coefficient, selecting at least one object for recommendation. Therefore, recommendation is carried out based on the user historical data and the object popularity, the method does not depend on a fixed model, limitation of data types can be restrained, adaptability in different types of objects is improved, the object popularity is represented through penalty factors, the obtained matching coefficient can be used for preferentially recommending the unpopular objects with low popularity, and the method is suitable for unpopular object recommendation scenes.

Description

technical field [0001] The present invention relates to the field of information technology, in particular to an object recommendation method and device, electronic equipment and a storage medium. Background technique [0002] In the purchase scene of goods, services and other objects, there is a phenomenon that the cumulative total revenue of unpopular objects with small sales volume but many types exceeds that of mainstream objects due to their huge total amount, which is called the long tail effect. At present, the long-tail product recommendation scheme under the long-tail effect is recommended by using the graph attention network, which needs to use the trained network model. The models in this type of scheme have a strong dependence on the data type, and the results obtained by the same network model in different data fields may vary greatly, resulting in poor transferability and adaptability of the fixed model between different types of data. Contents of the inventi...

AI Technical Summary

Problems solved by technology

The models in this type of scheme have a strong dependence on data types, and the results obtained by the same network model in different data fields may vary greatly, resulting in poor transferability and adaptability of fixed models between different types of data.

Images