摘要
针对传统网络资源缺失信息碎片识别方法中识别准确度较低、完成时间较长、能量消耗较大等问题,提出一种基于大数据分析的网络资源缺失信息碎片识别方法。通过对网络资源信息分析,利用非线性时间序列对网络资源不完整信息进行相空间重建,引入关联维数对网络资源不完整信息特征提取;考虑到不完整信息特征中缺失信息碎片对信息类别的贡献度,利用信息熵来衡量缺失信息碎片之间的差异,利用以BP神经网络为基础的集成分类器对缺失信息碎片分类,完成缺失信息碎片识别。结果表明,所提方法识别准确度较高、完成时间较短、能量消耗较小。
Aiming at the traditional network resource missing information fragment identification method, there are generally problems such as low recognition accuracy, long completion time and large energy consumption. This paper proposes a method for identifying missing information fragments of network resources based on information entropy and integrated classification. By analyzing the network resource information, the nonlinear spatial time series is used to reconstruct the incomplete information of the network resources, and the correlation dimension is introduced to extract the incomplete information features of the network resources, taking into account the information of the missing information in the incomplete information features. The contribution of categories, using information entropy to measure the difference between missing information, the BP neural network-based integrated classifier classifies the missing information and completes the identification. The experimental results show that the proposed method has higher recognition accuracy, shorter completion time and less energy consumption.
作者
李田英
LI Tian-ying(Modern Education and Technology Center/Shangqiu Medical College, Shangqiu 476100, China)
出处
《山东农业大学学报（自然科学版）》
北大核心
2019年第5期870-872,共3页
Journal of Shandong Agricultural University：Natural Science Edition
基金
2015年河南省医学教育研究项目:依托网络专题教育社区的医学超声诊断技术教学模式改革的探索(Wjlx2015170)
关键词
大数据分析
网络资源
缺失信息
智能识别
Big data analysis
network resource
missing information
intelligent identification
