Data analysis method and device

Abstract

Embodiments of the invention disclose a data analysis method and device. The method comprises the steps of obtaining user data; dividing the user data into feature data and behavior data; performing data preprocessing on the feature data to obtain a first data set; carrying out feature engineering processing on the first data set to obtain a second data set, wherein the second data set comprises key features and key feature values corresponding to the key features; importing behavior data into the behavior prediction model to predict the behavior score of the user; determining key features of the user and key feature values corresponding to the key features by using the second data set; and establishing a user scoring table by using the key features, the key feature values corresponding to the key features and the behavior scores of the users. By adopting the method provided by the embodiment of the invention, the key features of different users and the corresponding feature values can be obtained, the behavior of the users can be obtained through the behavior prediction model, and the key features of the clients can be dynamically displayed according to the radar map to perform user business risk control.

Description

technical field [0001] The present application relates to the field of data processing and analysis, in particular to a data analysis method and device. Background technique [0002] In the current digital marketing scenario, marketers can view user portraits to understand the user's usage on each function page of the mini program or APP, and can also calculate user labels through preset rules and display them on user portraits. Machine learning modeling can also be used to delineate user groups that may perform a certain target behavior, such as high-churn and high-intention users, and carry out retention or incentive marketing for users. [0003] User portraits in existing solutions require business personnel or data analysts to establish rules and algorithms through a large amount of data analysis to form corresponding labels, and the labor cost is huge; the generated rule labels and statistical labels, marketers can only perceive the behavior of users , but it is imposs...

AI Technical Summary

Problems solved by technology

[0003] User portraits in existing solutions require business personnel or data analysts to establish rules and algorithms through a large amount of data analysis to form corresponding labels, and the labor cost is huge; the generated rule labels and statistical labels, marketers can only perceive the behavior of users , but it is impossible to gain insight into which behaviors of users are conducive to their own business, and it is also impossible to make effective choices from tags of many dimensions; tags are often only applicable to one scenario, and marketers cannot determine whether the tag is useful for the current scenario

Images