TC: Medium: Security and Privacy Preserving Data Mining and Management for Disctributed Domains

在信息安全方面的一个基本但具有挑战性的问题是在形成大规模电子企业的众多组织之间安全地共享和管理敏感数据和信息。今天，越来越多的企业正在使用互联网管理和共享用户？和企业信息。然而，数据的安全和隐私是目前限制信息技术扩散的首要问题。除其他外，两个关键的安全问题是：（a）协作组织为共享敏感信息而实行的各种安全政策的互操作性挑战;（B）利用协作知识来检测和应对任何新出现的安全或紧急威胁。本项目的目标是开发一个可扩展的多域信息安全框架，该框架：（a）促进驻留在多个域中的多媒体数据的大规模集成、挖掘和分析，以识别需要受控传播的敏感信息（B）允许跨多个域的多媒体数据流的分布式实时内容分析，以关联可检测事件，（c）l提供集成和发展的域间访问控制策略的分布式多媒体应用，和（d）确保保护用户？的信息。 该项目采用了几种创新方法来实现这些目标。首先，新的实时数据挖掘和流相关的方法被用于提取有用的内容和检测事件嵌入在数据流中，通过使用多个滑动窗口算子。第二，一个有效的隐私/安全为基础的多媒体数据聚类机制允许一个流行的安全策略，以适应不断变化的上下文和数据集。这种方法的独特之处在于安全策略的自适应性。此外，本研究亦采用一种崭新的以XML为基础的整合机制，结合多领域的安全策略，以确保系统的技术可行性及可移植性。这项研究将对发展安全的资讯基建，例如电子商贸、数码图书馆及医疗系统，预计未来几十年将在全球范围内发挥重要作用。研究结果可以帮助减轻与信息和系统安全相关的严重金融风险，并在总体上满足国家信息基础设施安全需求和威胁管理。 该研究将与网络中心，信息保障和安全教育和研究中心（CERIAS）和普渡大学的PURVAC设施合作进行，PI？有很强的联系和持续的合作。研究结果和开发的工具将通过网络提供给研究界。在该小组有一个跟踪记录，提供他们的研究成果和工具，以各种研究团体在多媒体数据库和信息安全管理的领域。欲了解更多信息，请参阅该项目的网站在网址：http://cobweb.ecn.purdue.edu/~dmultlab/ http://cobweb.ecn.purdue.edu/%7Edmultlab/

A fundamental but challenging issue in information security is secure sharing and management of sensitive data and information among numerous organizations that form large-scale e-enterprises. Today, an increasing number of enterprises are using the Internet for managing and sharing users? and enterprise information through online databases. However, security and privacy of data is an overriding concern currently limiting the proliferation of information technology. Among others, the two key security issues are; (a) interoperability challenge of diverse security policies exercised by collaborating organizations for sharing sensitive information and; (b) using collaborative knowledge for detecting and responding to any emerging security or emergency threats. The objective of this project to develop a scalable multi-domain information security framework which: (a) facilitates large-scale integration, mining, and analysis of multimedia data residing in multiple domains for identifying sensitive information that require controlled dissemination (b) allows distributed real-time content analysis of multimedia data streams across multiple-domains for correlating detectable events, (c)l provides integration and evolution of inter-domain access control policies for distributed multimedia applications, and (d) ensures protection of user?s information. The project uses several innovative approaches to meet these objectives. First, novel real-time data mining and stream-correlation methods is used for extracting useful content and detecting events embedded in data streams by using multiple sliding window operators. Second, an efficient privacy/security based multimedia data clustering mechanism allows a prevailing security policy to be adaptable to the changing context and the datasets. The uniqueness of this approach is the adaptive nature of security policies. In addition, a novel XML-based integrated mechanism is used for encoding of multimedia data in conjunction with multi-domain security policies for ensuring technical viability and portability of the system.The proposed research is expected to have direct and long-term impact on developing secure information infrastructures, such as e-commerce, digital libraries, and healthcare systems, which are projected to have an important role at a global scale for the next several decades. The research results can help in alleviating the heavy financial risks associated with information and system security and in general, meeting the national information infrastructure security needs and threat management. The research will be pursued in collaboration with Cyber Center, the Center for Education and Research in Information Assurance and Security (CERIAS) and PURVAC facilities at Purdue University with which the PI?s have strong affiliation and on-going collaboration. Results and tools developed will be made available over the Web to research community. In The team has a track record of providing their research results and tools to various research communities in the areas of multimedia databases and information security management.For further information see the project web site at theURL: http://cobweb.ecn.purdue.edu/~dmultlab/ http://cobweb.ecn.purdue.edu/%7Edmultlab/