大数据下网络资源信息丢失优化识别仿真

发布时间：2018-05-04 04:42

本文选题：大数据 + 网络资源　；参考：《计算机仿真》2017年09期

【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。
[Abstract]:The optimization identification of network resource information loss under large data can ensure the stable and normal operation of the network. The recognition of the lost resource information needs to get the new data fragments from the concentrated point, thus training the decision function to complete the information of the network resource loss. The traditional method obtains the characteristics of the network resource information combination. Vector, the information combination feature vector is lost recognition, but the decision function is ignored, and the recognition accuracy is low. Based on the fuzzy C mean clustering, the network resource information loss recognition algorithm under large data is proposed, and the different types of network resource information samples are collected under large data, and the samples of different types of network resource information are carried out. Feature extraction, using the fuzzy C mean clustering theory to cluster analysis of debris samples, using information concentration criteria to deal with the cluster center of debris samples, get the new data fragments composed of concentrated points, and use the new data fragments to train to obtain the number of decision functions, in order to complete the loss of information of network resources. The experimental results show that the proposed algorithm can effectively improve the recognition accuracy of network resource information and has strong practicability.

【作者单位】：四川大学锦城学院;
【基金】：四川省教育厅重大培育项目(16CZ0039)
【分类号】：TP311.13;TP393.0

【相似文献】