Merchant popularity calculation method and device, electronic equipment and readable storage medium

Abstract

The invention discloses a merchant popularity calculation method. The merchant popularity calculation method comprises the steps: obtaining user behavior data related to target merchant popularity; based on the user behavior data, obtaining popularity characteristics of the target merchant; and according to the popularity characteristics and the popularity characteristic weight corresponding to each popularity characteristic, obtaining a popularity value of the target merchant, wherein the heat characteristic weight is a numerical value obtained through an unsupervised mode based on user behavior data of a plurality of sample merchants. Due to the merchant popularity calculation method, the technical problems that the objectivity of the calibration result of the sample data of the heat degree determination scheme is insufficient, the feature weight is not easy to determine and the accuracy is poor are solved. The merchant popularity calculation method has the beneficial effects that the objectivity of the sample data is improved, and the popularity characteristic weight and the popularity evaluation result accuracy are improved.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a merchant popularity calculation method, device, electronic equipment and readable storage medium. Background technique [0002] With the advent of the "big data" era, more and more companies have begun to mine the commercial value of data and use data heat for real-time digital marketing. At present, in order to avoid the problem of high pressure on the server caused by the user requesting the server to search for product information in a large amount of product information, the commonly used solution is mainly to predict the popularity of the product and provide product information to the user according to the prediction result. [0003] Disadvantages of existing heat calculation related technologies: On the one hand, the weights of heat-related feature factors need to be constructed in a data training set and learned through a supervised model. In reality, sample da...

AI Technical Summary

Problems solved by technology

[0003] Disadvantages of existing heat calculation-related technologies: On the one hand, the weight of heat-related feature factors needs to be constructed in a data training set and learned through a supervised model. In reality, sample data needs to be manually calibrated, which is highly subjective and thus easy affect the accuracy of calibration results; on the other hand, if the weights of heat-related feature factors are formulated according to rules, the rules need to be determined based on certain manual experience. When there are many features or complex conditions, it is difficult to determine feature weights based on rules and less accurate

Images