Privacy Data Measurement and Classification Model Based on Shannon Information Entropy and BP Neural Network

• Published in: 2023 International Conference on Data Science and Network Security (ICDSNS) pp. 1 - 5
• Main Author: Wang, Chengsi
• Format: Conference Proceeding
• Language: English
• Published: IEEE 28.07.2023
• Subjects: BP neural network; Current measurement; Data models; Data privacy; Data protection; Machine learning algorithms; measurement classification; Neural networks; Privacy; privacy data; shannon information entropy
• DOI: 10.1109/ICDSNS58469.2023.10244966

The protection of privacy information is an important issue in modern network research. How to effectively protect a large number of incorrect, unreasonable and unauthorized personal privacy information and data. In this paper, based on Shannon entropy and BP neural algorithm, a classification effect model based on statistical learning method is constructed. After getting the results under different parameter conditions, we can take them as reference objects. According to the model, a classification function of privacy information protection policy with good classification performance is established, and the algorithm and implementation method that meet the above research requirements are designed using the network structure. The test results show that the overall accuracy of the data protection measurement and classification model proposed in this paper exceeds 96%, and can also provide more than 95% accuracy for a single sample of data protection level. On the other hand, we can see from the error estimation rate of each data protection level that the classification accuracy of the data model under the relative data protection limit is higher than that of the data model under the intermediate data protection level. The greater the difference between the personal data provided by this data, the easier it is to measure privacy.