摘要
大数据下网络资源信息丢失的优化识别,能够保证网络稳定正常运行。对丢失资源信息的识别,需要得到浓缩点组成的新数据碎片样本,由此进行训练获得决策函数,完成网络资源丢失信息的识别。传统方法获得网络资源信息组合特征向量,对信息组合特征向量进行丢失识别,但忽略了获取决策函数,导致识别精度偏低。提出基于模糊C均值聚类的大数据下网络资源信息丢失识别算法,采集大数据下不同类型的网络资源信息样本,对不同类型的网络资源信息样本进行特征提取,通过模糊C均值聚类理论对碎片样本进行聚类分析,利用信息浓缩准则对碎片样本聚类中心进行处理,得到浓缩点组成的新数据碎片样本,并使用新数据碎片样本进行训练获得决策函数,以此为依据完成网络资源信息丢失识别。实验结果表明,所提算法能够有效提高网络资源信息识别精度,实用性较强。
An identification algorithm for information loss of network resource under big data is proposed based on fuzzy c-means cluster. Firstly, different types of information samples of network resource under the big data are collected and the feature for the information samples is extracted, then clustering analysis for fragment sample is carried out via the fuzzy c-means duster. Moreover, the criterion of information concentration is used to process center of clustering of the fragment sample, the fragment sample of new data with concentration point is obtained, the fragment sample is used to carry out training, and the decision function is obtained. Finally, the identification of information loss is completed. Experimental results show that the algorithm can improve precision of the identification.
出处
《计算机仿真》
北大核心
2017年第9期358-361,共4页
Computer Simulation
基金
四川省教育厅重大培育项目(16CZ0039)
关键词
大数据
网络资源
丢失信息识别
Big data
Network resource
Identification of loss information
