Shopping behavior prediction method and device

Abstract

The invention discloses a shopping behavior prediction method, and the method comprises the steps: selecting different target users at different shopping stages, and obtaining sample data from a user behavior log of the selected target users; respectively extracting a first feature set for marking user behavior from the sample data at each shopping stage; respectively training the first feature sets at different stages through a decision tree model, obtaining a feature combination through multiple iteration, and enabling the feature combination to serve as a second feature set; respectively employing the first and second feature sets at all shopping stages to train a pre-built machine learning model, wherein the machine learning model is used for predicting the shopping demand degree of a user; and determining the shopping stage of a to-be-detected user according to the machine learning model at different shopping stages. The invention also provides a corresponding shopping behavior prediction device.

Description

technical field [0001] The invention relates to the Internet field, in particular to a method and device for predicting user shopping behavior. Background technique [0002] The prediction of users' shopping needs for specific commodities in the future is of great significance for Internet companies, especially e-commerce companies, to make personalized business decisions and refined advertising decisions. In addition, in many application scenarios, it is also necessary to estimate the user's shopping needs in real time: for example, high-quality traffic sales in personalized advertising, volume-guaranteed CPM (Cost Per Mille, cost per thousand people) advertising strategy, especially e-commerce When the traffic of commercial websites increases, there is an urgent need for personalized shopping guide strategies and market positioning for users, so as to improve users' shopping fun and satisfaction. This is a difficult problem for Internet companies with massive data. [00...

AI Technical Summary

Problems solved by technology

In terms of algorithm implementation, one implementation method is to use the similarity analysis method to estimate the user's shopping intention, such as using the user's historical preference information to calculate the distance between users, and then using the target user's nearest neighbor user's evaluation of the product. The weighted evaluation value is used to predict the target user's preference for a specific product, but this method will have sparsity and scalability problems, and it is not suitable for the user's estimation of the purchasing power of a specific product.

Another implementation method is to use the linear model (LR) for modeling and estimation. The efficiency of the linear model is relatively high, but the linear model has a poor grasp of the nonlinear relationship, and the linear model cannot go deep into the details of the data. The selection of features It is very troublesome to extract and extract, and requires a lot of knowledge in related industries, so it is difficult to control the accuracy of the prediction model

Images