Abstract:

Facing the challenges of data security governance in the new situation and following the technological development trends in the field of data security, in response to the prominent issue of identifying key authorized personnel in the practical application of data security governance in large state-owned enterprises, this article proposes a key authorized personnel identification technology based on graph algorithm, which can discover potential authorization influencing factors in the system and measure the weight influence of different meanings from multiple perspectives, The recognition results have strong interpretability.Aiming at the problems of user and entity behavior anomaly detection in data security governance, this paper proposes a user and entity behavior anomaly detection method based on the generative adversarial network.The experimental results show that the accuracy, recall rate and average F1-score of the proposed method are better than the comparison baseline model method.A data security platform has been designed and developed.The platform has played an important role in reducing data security risks, assisting enterprise compliance construction, promoting data development and utilization, and has been applied in multiple data centralized management projects.It can meet the needs of big data processing in security scenarios, and has good application and promotion value.

Key words: data security governance, graph algorithm, UEBA, data security platform