基于个人敏感信息的数据脱敏技术研究

In the new age, it is the core production factor that data is flowing, shared and open among various industries. Due to the non-competitive nature of data, data security risks such as data leakage in the flow and openness are becoming more and more prominent. To address the problem that existing sensitive data recognition methods are difficult to scale and have low recognition accuracy and recognition performance, this paper proposes a sensitive data recognition method (abbreviated as: SDICBIA) based on classification first and recognition later architecture. It is shown that the recognition accuracy of SDICBIA can reach 97.14% and the recognition performance can be improved by about 70% compared with the traditional regular matching-based sensitive data recognition method. Secondly, in this paper, we design a data masking system that supports automatic and intelligent recognition of sensitive data as well as automatic data masking and can be used to prevent leakage of sensitive personal information. It mainly includes three functional modules: synchronous adaptation management module of data sources with multiple heterogeneous sources, automatic identification module of sensitive data, and data masking module. In which, the data masking system is designed and implemented for structured data application scene.