Prediction method and device for secondary purchase intention of customers

Abstract

The invention relates to the technical field of data processing, and discloses a prediction method and device for the secondary purchase intention of customers, used for improving the accuracy of a predicted result of the secondary purchase intention of the customers. The method comprises the following steps of: reading a target customer list from a specified storage position, acquiring historical benefits of respective customers according to customer identifiers stored in the obtained target customer list, determining the variation tendency of the historical benefits as well as the purchasing power parameters and customer maturity of the respective customers according to the historical benefits; calculating the user psychological comfort levels of the respective customers according to the acquired parameters; and generating a final marketing customer list according to customer identifiers of the customers whose user psychological comfort levels reach set conditions. Thus, the calculation workload can be reduced, the calculation speed can be improved, and therefore the operation pressure of a server can be effectively reduced. The invention also discloses an evaluation device.

Description

technical field [0001] The present application relates to the technical field of data processing, in particular to a method and device for predicting a customer's repurchase intention. Background technique [0002] With the development of technology, e-commerce has gradually become one of the important modes of enterprise operation. For enterprises, how to make customers repurchase products / services (including: cross-selling, up-selling, renewal sales) is a key issue. It is the key to enhance customer value and maintain the sustainable profitability of enterprises. The so-called re-purchase refers to continuing to purchase the same or other products / services after purchasing a product / service from a certain merchant. The so-called cross-selling is to sell more other products / services to customers who have already purchased a certain product / service; the so-called up-selling is to sell similar products / services of higher value to customers who have already purchased a certai...

AI Technical Summary

Problems solved by technology

[0003] At present, the predictive analysis of customers' repurchase behavior generally adopts the predictive analysis method of data mining, that is, customers who have already repurchased behaviors are used as sample customers, and predictive analysis methods such as decision trees and logistic regression are used to establish predictive analysis models. To predict the repurchase intention of other customers, this prediction method is affected by factors such as the number of sample customers and the stability of characteristics, and the validity of the prediction results is often difficult to guarantee; and because there are no sample customers to refer to when a new product is just launched, Conventional data mining predictive analysis methods will not be able to implement

[0004] When conventional data mining predictive analysis methods are applied to the Internet industry for mining and analysis of massive data (such as tens of millions, hundreds of millions of customer data), due to the huge amount of data and the complexity of the calculation process, the requirements for system resources are high and the calculation time is long. , it is difficult to meet the business requirements of the rapid response of the Internet industry

Images