基于DTS的多模态异构大数据检测方法研究

Research on multi-modal heterogeneous big data detection method based on DTS

为降低由多模态异构数据造成的信息传输风险,增强大数据网络对于信息攻击行为的实际承载能力,提出基于DTS的多模态异构大数据检测方法。利用深度置信网络对异构数据进行编码处理,再借助DTS线性映射条件,实现多模态异构大数据提取。在此基础上,清洗大数据环境下的所有传输信息参量,通过二值化转换的方式控制异构数据的降维方向,实现多模态异构大数据检测。对比实验结果表明,与传统检测方法相比,DTS检测方法可规避95%以上的风险性信息传输行为,且能使多模态异构数据的稳定抵抗能力得到大幅提升,满足增强大数据网络对于信息攻击行为承载能力的实际应用需求。

In order to reduce the risk of information transmission caused by multi-modal heterogeneous data and enhance the actual carrying capacity of big data network for information attack behavior,a multi-modal heterogeneous big data detection method based on DTS is proposed.The deep confidence network is used to encode and process heterogeneous data,and then the DTS linear mapping condition is used to realize multi-mode heterogeneous big data extraction.On this basis,all the transmission information parameters in the big data environment are cleaned,and the dimension reduction direction of heterogeneous data is controlled by means of binary conversion,so as to realize multi-mode heterogeneous big data detection.The experimental results show that compared with traditional detection methods,DTS detection method can avoid more than 95%of risky information transmission behavior,and can greatly improve the stable resistance ability of multi-modal heterogeneous data,so as to meet the practical application requirements of enhancing the carrying capacity of information attack behavior in big data network.