Item recommendation method, meta-network processing method, device, storage medium and equipment

Abstract

The present application relates to an item recommendation method, a meta-network processing method, a device, a storage medium and a device. The item recommendation method includes: obtaining attribute information of the item and an object identifier of an item interactive object; performing feature extraction on the attribute information through a meta-stretching network to obtain a stretching vector; performing feature extraction on the logo to obtain an offset vector; performing feature crossing between the vectorized item logo of the item and the stretch vector, and fusing the obtained cross feature vector with the offset vector to obtain a fusion feature vector ; determining a recommendation probability based on the fusion feature vector, the vectorized attribute information and the vectorized object information of the target object; recommending the item to the target object when the recommendation probability meets a recommendation condition. Using this method can improve the accuracy of item recommendation, so as to achieve the effect of accurate recommendation.

Description

Technical field[0001]This application relates to the field of artificial intelligence technology, in particular to an item recommendation method, meta-network processing method, device, storage medium, and equipment.Background technique[0002]With the continuous development of Internet technology, users can obtain various items from the Internet, such as applications or other service products. For item providers, it is necessary to accurately recommend new items developed to users in need to increase product utilization.[0003]Since new items developed are usually only used by a small number of users, when making recommendations, the traditional recommendation scheme not only recommends users interested and mature items, but also pushes the new items to users to improve the use of new items rate. However, in the above recommendation scheme, the user is interested in mature items, but not necessarily interested in the new items, resulting in low accuracy in recommending items to the us...

AI Technical Summary

Problems solved by technology

However, in the above recommendation scheme, the user is interested in the mature item, but not necessarily interested in the new item, resulting in low accuracy in recommending items to the user

Images