In the field of education,particular attention is paid to data security issues in cloud platforms,as they are directly re-lated to the protection of the privacy of students and teachers and the secure storage of teaching content.In this paper,the multi-label kernel mapping data dimensionality reduction method and least squares multi-classification twin support vector machine model are ap-plied,and the invasive weed optimization algorithm is combined.The experimental results show that the mining error is about 2%,and the energy saving is between 400J-500J,which is higher than the other two methods.The maximum mining time was 2.2 seconds,much faster than the 7.8 seconds and 8.4 seconds of other methods.The results show that the proposed method has obvious advantages in dealing with advanced persistent threat data on cloud computing platforms.
关键词
云平台/数据挖掘/孪生支持向量机/入侵杂草优化算法
Key words
cloud platform/data mining/twin support vector machine/invasive weed optimization algorithm
维普
万方数据