大数据网络信息系统缺失优化检测仿真研究

发布时间：2018-05-16 01:38

本文选题：网络信息系统 + 数据缺失　；参考：《计算机仿真》2017年09期

【摘要】：对大数据网络信息系统的数据缺失的有效检测,能够提升对大数据分析的完整性和全面性。对数据缺失值的检测,需要提取每个数据属性的深度特征,进行缺失值的预测,完成对数据缺失值进行检测。传统方法度量不同数据缺失样本的相关性,构造数据缺失样本的相似度矩阵,但忽略了对缺失值进行预测,导致检测精度低。提出基于支持向量机的大数据网络信息系统数据缺失值检测方法。融合于深度学习理论将检测自动编码机定义为基础模块,从完整数据子集中提取不完全数据集中每个数据属性的深度特征,利用支持向量机方法,将缺失值检测划分为连续、类别属性缺失值检测两种情况,进行缺失值的预测,并完成了对大数据网络信息系统数据缺失值的检测。仿真证明,所提方法检测精度高,能够有效检测缺失数据。
[Abstract]:The effective detection of the missing data in big data network information system can improve the integrity and comprehensiveness of big data analysis. To detect the missing value of the data, we need to extract the depth feature of each data attribute, predict the missing value, and complete the detection of the missing value of the data. The traditional method measures the correlation of different missing data samples and constructs the similarity matrix of missing data samples, but neglects to predict the missing values, which results in low detection accuracy. This paper presents a method for detecting the missing values of big data network information system based on support vector machine (SVM). Based on the theory of depth learning, the detection automatic coding machine is defined as the basic module, and the depth features of each data attribute in the incomplete data set are extracted from the complete data subset. By using the support vector machine method, the missing value detection is divided into continuous. In this paper, the missing value of big data network information system is detected in two cases, and the missing value is predicted. Simulation results show that the proposed method has high accuracy and can effectively detect missing data.
【作者单位】：中原工学院计算机学院;
【基金】：基金项目:OAIS的电子文件生命周期可信安全管理平台(15A520113)
【分类号】：TP311.13;TP393.0

【相似文献】