Classification and grading method for sensitive data in digital power grid based on hybrid feature ranking algorithm

Le résumé fourni par la source

Traditional data classification algorithms have some limitations when dealing with sensitive data in digital power grids. Therefore, this paper studies the classification and categorization of sensitive data in digital power grids based on hybrid feature classification algorithms. Extract mixed features of digital grid sensitive data from three dimensions: volatility, trend, and variability. The support vector machine classifier optimized by the krill algorithm is used to classify the sensitive data of the digital grid after feature dimensionality reduction, and the K-means clustering algorithm is selected to cluster various sensitive data of the digital grid, obtaining the classification and classification results of the sensitive information of the digital grid. The experimental results show that this method can accurately classify and grade sensitive information of the digital power grid, with a classification accuracy rate of over 98%. Based on data classification and grading results, the protection level of sensitive data in the digital power grid has been improved.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.