An Efficiency Optimization Study of Data Governance Legal Issues in the Framework of Privacy Computing

The rapid development of technologies such as big data and cloud computing provides convenience for people’s production and life but also brings hidden dangers for personal privacy. The study clarifies the legal norms and guarantees of personal information protection (privacy) in countries around the world and, on the basis of this law, adopts the key technologies of privacy computation, such as differential privacy, machine learning, and homomorphic encryption to protect the privacy of information data with symmetric encryption, and constructs a privacy protection computational model based on symmetric encryption. To evaluate the efficiency of this paper’s privacy protection computing model, it is necessary to compare its performance with other privacy computing model algorithms. The output of this paper’s privacy-preserving model is stable on the three datasets, DBLP, PUB, and CITESEERX, with minimal result perturbation and good accuracy performance. The accuracy of this paper’s algorithm at θ =0.5 is 99.8% and 96.8%, respectively, and balances the training efficiency and accuracy. The running time of this paper’s model on DBLP and PUB datasets is 47.5 minutes and 39.6 minutes, respectively, which is more efficient. The privacy-preserving computational model in this paper significantly outperforms other algorithms in terms of system throughput and improves the efficiency of the system’s data processing while reducing the consumption and waste of arithmetic resources.