Commodity pushing method and apparatus for online transaction

Abstract

Embodiments of the invention disclose a commodity pushing method and apparatus for online transaction, which relates to the technical field of the internet and can improve the accuracy of commoditiesrecommended to users. The method comprises the steps of reading to-be-processed order data of commodities which correlation rules point to, wherein the correlation rules are used for representing at least two commodities with a correlation relationship; according to the order data of the commodities which the correlation rules point to, obtaining a support degree and a confidence degree of the correlation rules, wherein the support degree is used for representing a conditional probability of the commodities which the correlation rules point to in date granularity, and the confidence degree isused or representing a conditional probability of the commodities which the correlation rules point to in sales volume; according to the support degree and the confidence degree of the correlation rules, filtering invalid rules from the correlation rules; and according to the filtered correlation rules, determining to-be-recommended commodities, and pushing the to-be-recommended commodities to user equipment. The commodity pushing method and apparatus is suitable for commodity pushing in related online transactions during shopping of the users.

Description

technical field [0001] The invention relates to the technical field of the Internet, in particular to a method and device for pushing commodities in online transactions. Background technique [0002] The current major online shopping platforms have adopted some recommendation and promotion schemes around product sales, so as to recommend related products when users are shopping, thereby increasing the transaction volume. It mainly analyzes the correlation between the products purchased by the user to obtain the recommendation rules, and after the user completes a purchase behavior, actively recommends related products according to the recommendation rules, thereby increasing the cross-selling opportunities of the products. [0003] However, at present, association rule analysis is still based on the granularity of the actual accumulated order quantity. Once encountering non-daily sales fluctuations such as promotional activities and popular products, the proportion of such p...

AI Technical Summary

Problems solved by technology

[0003] However, at present, association rule analysis is still based on the granularity of the actual accumulated order quantity. Once encountering non-daily sales fluctuations such as promotional activities and popular products, the proportion of such products in the association rules will be too high. However, these categories Commodities often have a large market demand and cannot be blocked in the process of mining association rules

And finally, because the sales volume of each product varies greatly, and the external factors that affect the sales volume of the product are very complicated in practice, the accuracy of the product recommended to the user is low, and it is difficult to meet the real needs of the user.

Images