Cross-domain data fusion method, system and storage medium

Abstract

A cross-domain data fusion method and system, and a storage medium, relating to the technical field of data mining. The methods comprises: obtaining user feature information corresponding to user keyword information stored in a data source in a data domain (101); converting the user feature information into implicit feature information on the basis of a feature conversion model (102); encrypting the user keyword information according to a preset encryption rule (103); fusing the user keyword information cipher text and the implicit feature information to generate a domain user fusion feature of the data domain (104); and fusing the domain user fusion features of multiple data domains according to the user keyword information cipher text to obtain a cross-domain user fusion feature (105). The cross-domain data fusion method and system and the storage medium can avoid user original data leakage caused by the leakage of user keywords, converting the user feature information into the implicit feature information and performing fusion enhance the expression ability of the data aggregation features in different fields, and the accuracy and security of data mining can be improved.

Description

technical field [0001] The present disclosure relates to the technical field of data mining, and in particular to a cross-domain data fusion method, system and storage medium. Background technique [0002] Traditional data mining usually deals with data from a single data domain, merging multiple data sets in the same domain into a database with a consistent data model. In the context of the big data era, we are often faced with various types of data from different fields and sources. These data are often composed of multiple forms, and each form has different representations, distributions, scales, and densities. Multiple datasets from different domains are implicitly connected through latent objects, for example, a user’s e-commerce platform transaction data, personal credit history, and demographic characteristics describe the user’s potential spending power. [0003] Existing cross-domain data fusion first performs user feature fusion in different storage sources, and t...

AI Technical Summary

Problems solved by technology

The user's data in different fields is privacy-sensitive. Once the encryption method of the user Key is leaked, the attacker can use the ciphertext of the user Key to reversely obtain the user Key, resulting in the leakage of the user's knowledge in various fields and even the original feature information, and each storage location The output is composed of user Key ciphertext and low-level original feature fusion, which is less secure

Images