Commodity recommendation method, device, electronic commerce system and storage medium

Abstract

The invention discloses a commodity recommendation method, device, e-commerce system and storage medium, and relates to the technical field of e-commerce. The method includes: receiving a commodity recommendation request sent by a client, and the information carried in the commodity recommendation request includes: commodity category, The first user location information and the number of target groups N; based on the first user location information, N target users are screened out, and the commodities purchased by the target users include candidate commodities that match the commodity category; according to the recommendation rules, select from the candidate commodities The recommended products are screened out, and the product recommendation information corresponding to the recommended products is sent to the client. The method, device, e-commerce system and storage medium of the present invention realize personalized recommendation based on the number of people around the user, the user can dynamically select the range of the number of people to be recommended, and can more flexibly obtain the commodities the user needs, improving the online The efficiency of purchasing goods improves the user's purchasing experience.

Description

technical field [0001] The present invention relates to the technical field of e-commerce, in particular to a method and device for recommending commodities, an e-commerce system and a storage medium. Background technique [0002] For the current e-commerce manufacturers, the recommendation system has become one of the important ways to sell products. Merchants will recommend similar products according to the user's preferences, and users will choose products purchased by users with the same attributes as themselves. One of the key attributes of the same attributes is geographical location. It is found through research that users who are usually located in the same location range have great similarity in product preferences. In the existing recommendation system, the geographical range is selected as an important attribute to recommend the relevance of products. After the user logs in to the e-commerce company, the recommendation system recommends products based on the use...

AI Technical Summary

Problems solved by technology

However, after the geographical range is set, for example, within a range of 5km or 10km from the user, there may be no relevant shopping behaviors or orders generated within this geographical range, or the number of people who have shopping behaviors or generate orders is small, which is less than the user's expected The number of people who have shopping behavior or generated orders, resulting in the inability to get accurate recommendation results

Images