互联网下图书馆大数据分析及协同异常检测平台研究

Research on Library Big Data Analysis and Collaborative Anomaly Detection Platform under the Internet

随着云计算技术的发展,图书馆管理工作已逐步趋向云化,然而由于该模式的动态和分布式架构的特点,系统安全性和隐私性是关键性指标,因此对于图书馆大数据分析应用,通过检测网络漏洞至关重要。本文提出了一种用于检测来自图书馆云计算服务应用环境中的网络攻击的协作异常检测框架(Collaborative Anomaly Detection Framework,CADF)。介绍了框架的功能和部署方式,同时框架在某图书馆的数据集上进行评估,以检查其在云计算环境中部署时的可信度。实验结果表明,该框架可以轻松处理大规模系统,因为其实施只需要估算网络观测的统计测量。

With the development of cloud computing technology,library management gradually becomes cloud-based.However,due to the dynamic nature of the model and the characteristics of distributed architecture,system security and privacy are key indicators,so for library big data,analyzing applications is critical by detecting network vulnerabilities.This paper proposes a Collaborative Anomaly Detection Framework(CADF)for detecting network attacks from the library cloud computing service application environment.It introduces the function and deployment of the framework,while the framework evaluates it on a library's dataset to check its credibility when deployed in a cloud computing environment.Experimental results show that the framework can easily handle large-scale systems because its implementation only requires estimating statistical measurements of network observations.