基于深度学习的云端大数据安全防护技术

Abstract

云端海量大数据是数据分析的基础，数据本身的安全性和准确性，对数据分析的结果有重要影响。针对云端大数据的特性，融合Hadoop的海量大数据处理以及数字水印相关技术，提出了一种以深度信念网络(DBN)作为智能分类的机制，通过对数据进行多层的训练和调整，对云端海量数据进行计算，得到其分布式表示，进而获取数据的篡改和判断的依据。实验表明，Hadoop和AI的结合，很好的实现了云端海量大数据的安全防护。 The cloud big data is the basis of the data analysis. The security and accuracy of the big data is es-sential to the result of data analysis. By combining Hadoop’s big data processing technology and digital watermarking technology, a classification with DBN as a smart strategy is proposed. The multilayer has been trained and adjusted by this scheme. The mass of data can be calculated and the distributed data can also be obtained which is the basis of the judgment of data tampering. The experiments show that the combination of Hadoop and AI is an effective method to the massive data security.