大数据安全与隐私保护研究

大数据时代的来临正影响着人们日常生活方式、工作习惯及思考模式.但目前数据在收集、存储和使用中面临着诸多风险,大数据所导致的隐私泄露为用户带来了困扰,虚假数据将导致错误或无效的大数据分析结果.该文分析了实现大数据安全与隐私保护所面临的技术挑战,整理了大数据安全与隐私保护关键技术,对我们信息安全领域的发展有一定的参考和借鉴意义.     

一种基于格的隐私保护聚类数据挖掘方法

由于云计算的诸多优势,用户倾向于将数据挖掘和数据分析等业务外包到专业的云服务提供商,然而随之而来的是用户的隐私不能得到保证.目前,众多学者关注云环境下敏感数据存储的隐私保护,而隐私保护数据分析的相关研究还比较少.但是如果仅仅为了保护数据隐私,而不对大数据进行挖掘分析,大数据也就失去了其潜在的巨大价值.本文提出了一种云计算环境下基于格的隐私保护数据发布方法,利用格加密构建隐私数据的安全同态运算方法,并且在此基础上实现了支持隐私保护的云端密文数据聚类分析数据挖掘服务.为保护用户数据隐私,用户将数据加密之后发布到云服务提供商,云服务提供商利用基于格的同态加密算法实现隐私保护的k-means、隐私保护层次聚类以及隐私保护DBSCAN数据挖掘服务,但云服务提供商并不能直接访问用户数据破坏用户隐私.与现有的隐私数据发布方法相比,论文的隐私数据发布基于格的最接近向量困难问题（CVP）和最短向量困难问题（SVP）,具有很高的安全性.同时算法有效保持了密文数据间距离的精确性,与现有研究相比挖掘结果也具有更高的精确性和可用性.论文对方法的安全性进行了理论分析并设计实验对提出的隐私保护数据挖掘方法效率进行评估,实验结果表明本文提出的基于格的隐私保护数据挖掘算法与现有的方法相比具有更高的数据分析精确性和更高的计算效率.     

位置大数据隐私保护研究综述

大数据时代移动通信和传感设备等位置感知技术的发展形成了位置大数据,为人们的生活、商业运作方法以及科学研究带来了巨大收益.由于位置大数据用途多样,内容交叉冗余,经典的基于“知情与同意”以及匿名的隐私保护方法不能全面地保护用户隐私.位置大数据的隐私保护技术度量用户的位置隐私,在信息论意义上保护用户的敏感信息.介绍了位置大数据的概念以及位置大数据的隐私威胁,总结了针对位置大数据隐私的统一的基于度量的攻击模型,对目前位置大数据隐私保护领域已有的研究成果进行了归纳.根据位置隐私的保护程度,可以把现有方法总结为基于启发式隐私度量、概率推测和隐私信息检索的位置大数据隐私保护技术.对各类位置隐私保护技术的基本原理、特点进行了阐述,并重点介绍了当前该领域的前沿问题：基于隐私信息检索的位置隐私保护技术.在对已有技术深入分析对比的基础上,指出了未来在位置大数据与非位置大数据相结合、用户背景知识不确定等情况下保护用户位置隐私的发展方向.     

基于迁移学习的多源数据隐私保护方法研究

隐私保护的多源数据分析是大数据分析的研究热点,在多方隐私数据中学习分类器具有重要应用。提出两阶段的隐私保护分析器模型,首先在本地使用具有隐私保护性的PATE-T模型对隐私数据训练分类器;然后集合多方分类器,使用迁移学习将集合知识迁移到全局分类器,建立一个准确的、具有差分隐私的全局分类器。该全局分类器无需访问任何一方隐私数据。实验结果表明,全局分类器不仅能够很好地诠释各个本地分类器,而且还可以保护各方隐私训练数据的细节。     

大数据环境中交互式查询差分隐私保护模型

随着大数据时代的到来,数据挖掘技术被广泛应用,而线性查询作为该技术中最基础和最频繁的操作,其隐私保护在数据分析和数据发布隐私保护中占有极其重要的位置。交互式线性查询的交互增加了数据的处理量,运用传统的隐私保护模型效率较低。为了解决大数据环境中交互式查询差分隐私保护问题,模型针对大规模数据集中交互式线性查询差分隐私保护的特点,通过数据关联性分析减少冗余信息,采用交替方向乘子法对查询负载矩阵进行分解,利用自适应加噪技术产生差分隐私保护所需要的合理数量的噪声,设计并行处理方法实现该模型的计算。实验将提出的模型与以往模型进行对比。结果表明,所提出的模型在提升隐私保护精度的同时,也极大地提高了算法性能,因此模型切实可行。     

ESA:一种新型的隐私保护框架

随着大数据驱动下智能技术的快速发展,大规模数据收集场景成为数据治理和隐私保护的主战场,本地化差分隐私技术作为该场景下的主流技术,被谷歌、苹果、微软等企业广泛使用.然而,该技术在用户本地对数据进行扰动,引入较多噪声,数据可用性较差.为实现可用性与隐私性兼顾的隐私保护方法,ESA(encode-shuffle-analyze)框架被提出,它在混洗器(shuffler)的作用下尽可能对数据进行较小扰动,同时保护用户隐私,使得任一用户的隐私信息都不能被数据分析者从收集数据中唯一识别.鉴于差分隐私在数学上优雅且严格的隐私定义,该框架目前主要基于差分隐私技术进行实现,该种实现称为混洗差分隐私(shuffle differential privacy, SDP).在保证相同隐私损失ε的情况下,混洗差分隐私比本地化差分隐私的可用性高■倍,接近中心化差分隐私而不依赖于可信第三方.为对该新型的隐私保护框架进行综述,首先对该框架进行分析;之后基于主流的混洗差分隐私技术,对相关理论基础与技术基础进行总结,对不同统计问题下的隐私保护机制进行理论与实验对比;最终提出ESA框架的挑战问题,并对该框架下非差分隐私方法...     

大数据隐私保护密码技术研究综述

大数据是一种蕴含大量信息、具有极高价值的数据集合.为了避免大数据挖掘泄露用户的隐私,必须要对大数据进行必要的保护.由于大数据具有总量庞大、结构复杂、处理迅速等特点,传统的保护数据隐私的技术很多都不再适用.从密码学的角度,综述了近年来提出的、适用于大数据的隐私保护技术的研究进展.针对大数据的存储、搜索和计算这3个重要方面,分别阐述了大数据隐私保护的研究背景和主要研究方向,并具体介绍了相关技术的最新研究进展.最后指出未来大数据隐私保护研究的一些重要方向.     

一种有效的差分隐私事务数据发布策略

近年来,隐私保护事务数据发布得到了研究者的广泛关注.事务数据的稀疏性导致个体隐私保护与数据效用性之间很难达到平衡.目前已有的方法大多是基于分组的匿名模型,但该类模型依赖于攻击者背景知识,且发布的数据无法满足事务数据分析任务的需要.针对事务数据隐私保护发布的数据安全性与效用性不足,基于差分隐私与压缩感知理论,提出一种有效的面向应用的事务数据发布策略(transaction data publish strategy,TDPS).首先构建事务数据库的完整Trie项集树,然后基于压缩感知技术对项集树添加满足差分隐私约束的噪音得到含噪Trie项集树,最后在含噪树上进行频繁项集挖掘任务.实验结果表明,TDPS不仅能很好地保护隐私,而且能有效保持数据效用性,满足事务数据分析任务对数据质量的要求.     

安全多方计算及其在机器学习中的应用

随着人工智能、大数据等技术的发展,数据采集、数据分析等应用日渐普及,隐私泄露问题越来越严重.数据保护技术的缺乏限制了企业之间数据的互通,导致形成"数据孤岛".安全多方计算(securemultiparty computation,MPC)技术能够在不泄露明文的情况下实现多方参与的数据协同计算,实现安全的数据流通,达到数据"可用不可见".隐私保护机器学习是当前MPC技术最典型也是最受关注的应用与研究领域,MPC技术的应用可以保证在不泄露用户数据隐私和服务商模型参数隐私的情况下进行训练和推理.针对MPC及其在隐私保护机器学习领域的应用进行全面的分析与总结,首先介绍了MPC的安全模型和安全目标;梳理MPC基础技术的发展脉络,包括混淆电路、不经意传输、秘密分享和同态加密;并对MPC基础技术的优缺点进行分析,提出不同技术方案的适用场景;进一步对基于MPC技术实现的隐私保护机器学习方案进行了介绍与分析;最后进行总结和展望.     

机器学习中差分隐私的数据共享及发布：技术、应用和挑战

近年来,基于机器学习的数据分析和数据发布技术成为热点研究方向。与传统数据分析技术相比,机器学习的优点是能够精准分析大数据的结构与模式。但是,基于机器学习的数据分析技术的隐私安全问题日益突出,机器学习模型泄漏用户训练集中的隐私信息的事件频频发生,比如成员推断攻击泄漏机器学习中训练的存在与否,成员属性攻击泄漏机器学习模型训练集的隐私属性信息。差分隐私作为传统数据隐私保护的常用技术,正在试图融入机器学习以保护用户隐私安全。然而,对隐私安全、机器学习以及机器学习攻击三种技术的交叉研究较为少见。本文做了以下几个方面的研究:第一,调研分析差分隐私技术的发展历程,包括常见类型的定义、性质以及实现机制等,并举例说明差分隐私的多个实现机制的应用场景。初次之外,还详细讨论了最新的Rényi差分隐私定义和Moment Accountant差分隐私的累加技术。其二,本文详细总结了机器学习领域常见隐私威胁模型定义、隐私安全攻击实例方式以及差分隐私技术对各种隐私安全攻击的抵抗效果。其三,以机器学习较为常见的鉴别模型和生成模型为例,阐述了差分隐私技术如何应用于保护机器学习模型的技术,包括差分隐私的随机梯度扰动(DP-SGD)技术和差分隐私的知识转移(PATE)技术。最后,本文讨论了面向机器学习的差分隐私机制的若干研究方向及问题。     

Business Analytics-Based Enterprise Information Systems

Big data analytics and business analytics are a disruptive technology and innovative solution for enterprise development. However, what is the relationship between business analytics, big data analytics, and enterprise information systems (EIS)? How can business analytics enhance the development of EIS? How can analytics be incorporated into EIS? These are still big issues. This article addresses these three issues by proposing ontology of business analytics, presenting an analytics service-oriented architecture (ASOA) and applying ASOA to EIS, where our surveyed data analysis showed that the proposed ASOA is viable for developing EIS. This article then examines incorporation of business analytics into EIS through proposing a model for business analytics service-based EIS, or ASEIS for short. The proposed approach in this article might facilitate the research and development of EIS, business analytics, big data analytics, and business intelligence.     

State of the art in big data applications in microgrid: A review

The prospering Big data era is emerging in the power grid. Multiple world-wide studies are emphasizing the big data applications in the microgrid due to the huge amount of produced data. Big data analytics can impact the design and applications towards safer, better, more profitable, and effective power grid. This paper presents the recognition and challenges of the big data and the microgrid. The construction of big data analytics is introduced. The data sources, big data opportunities, and enhancement areas in the microgrid like stability improvement, asset management, renewable energy prediction, and decision-making support are summarized. Diverse case studies are presented including different planning, operation control, decision making, load forecasting, data attacks detection, and maintenance aspects of the microgrid. Finally, the open challenges of big data in the microgrid are discussed.     

医疗大数据及隐私泄露

大数据中的医疗大数据与人类的健康生活息息相关，随着大数据的发展、信息化的加快，医疗卫生信息平台、数字化的医疗设备与仪器迅速普及，导致医疗领域内的数据呈爆炸式增长，且类型繁多、关系复杂。敏感的医疗数据安全问题同样备受关注。医疗数据在为人类的健康提供帮助的同时，保护相关联的敏感数据越来越成为学者、从业者和普通大众所关注的热点。本文从大数据的基本概念入手，通过对现阶段隐私泄露及医疗大数据的相关研究进行分析，结合大数据领域的相关研究对当前隐私泄露行为、保护技术等问题进行分类阐述，希望能为本领域学者的进一步研究有所启示和帮助。     

Optimizing the energy efficiency of chiller systems in the semiconductor industry through big data analytics and an empirical study

Big data analytics is playing a more and more prominent role in the manufacturing industry as corporations attempt to utilize vast amounts of data to optimize the operation of plants and factories to gain a competitive advantage. Since the advent of Industry 4.0, also known as smart manufacturing, big data analytics, combined with expert domain knowledge, is facilitating ever-greater levels of speed and automaticity in manufacturing processes. The semiconductor industry is a fundamental driver of this transformation; moreover, due to the highly complex and energy-consuming nature of the semiconductor manufacturing process, semiconductor fabrication facilities (fabs) can also benefit greatly from incorporating big data analytics to improve production and energy efficiency. This paper developed a big data analytics framework, along with an empirical study conducted in collaboration with a semiconductor manufacturer in Taiwan, to optimize the energy efficiency of chiller systems in semiconductor fabs. Chiller systems are one of the most energy-consuming systems within a typical modern fab. The developed big data analytics framework allows production managers to ensure that chiller systems operate at an optimized level of energy efficiency under dynamically changing conditions, while fulfilling the chilling demands. Compared to the commonly-used heuristics previously employed at the fab to tune chiller system parameters, by the utilization of big data analytics, it is shown that fabs can achieve substantial energy savings, greater than 12%. The developed framework and the lessons learned from the empirical study are not only generalizable but also useful for practitioners who are interested in applying big data analytics to optimize the performance of other equipment systems in fabs.     

大数据分析中的计算智能研究现状与展望

随着产业界和科学界数据量的爆炸式增长,大数据技术和应用吸引了众多的关注.如何分析大数据,充分挖掘大数据的潜在价值,成为需要深入探讨的科学问题.计算智能是科学研究和工程实践中解决复杂问题的有效手段,是人工智能和信息科学的重要研究方向,应用计算智能方法进行大数据分析具有巨大的潜力.对大数据分析中的计算智能方法进行综述,结合大数据的特征,讨论了大数据分析中计算智能研究存在的问题和进一步的研究方向,阐述了数据源共享问题,并建议利用以天文学为代表的数据密集型基础科研领域的数据开展大数据分析研究.     

生物医疗健康大数据应用支撑平台与关键技术

随着生物信息学的不断发展,生物医学领域积累了大量的数据,大数据已经贯穿基础研究、临床诊断、医药开发、健康管理等生物医学领域的各个环节。如何有效存储、管理、分析这些海量数据面临严峻的而挑战。基于超级计算机的计算分析和存储能力,在生物医学大数据处理的异构融合架构,面向生物医学大数据的层次式存储系统,生物医学大数据处理的异构并行计算和多源数据的汇聚机制与分析方法,突破生物医学大数据的汇聚、存储、分析等方面的关键技术,构建一个计算、分析处理和存储融合平台,以满足多种类型生物医学大数据应用的不同需求。     

Big Data Initiatives: Minimizing Risk and Maximizing Value with Governance

Abstract

Big Data Analytics can be a fantastic business opportunity for many organizations. Already organizations are using advanced analytics to streamline production processes, optimize back office activities, market more effectively, and better satisfy customer demand. That said, it goes without saying (as recent headlines can attest) that sometimes enhanced analytics capabilities can introduce risks such as erosion of privacy, overly-intrusive knowledge about customers, etc.

Given this dichotomy, making the decision about when, whether, how much, and how to invest in big data analytics initiatives can be a challenge. Invest too soon and you may obviate existing investments or disrupt business activities; invest too late and you may find that competitors gain advantages that make the market landscape asymmetric.

This article outlines how and why applying “tried and true” governance principles can help make this decision easier. For those that have formalized governance structures in place, how they might inform the decision an organization makes in this regard – and for those that don’t have a formalized governance program – how they might co-opt some of those principles to help make this decision more approachable.     

Manufacturing big data ecosystem: A systematic literature review

Advanced manufacturing is one of the core national strategies in the US (AMP), Germany (Industry 4.0) and China (Made-in China 2025). The emergence of the concept of Cyber Physical System (CPS) and big data imperatively enable manufacturing to become smarter and more competitive among nations. Many researchers have proposed new solutions with big data enabling tools for manufacturing applications in three directions: product, production and business. Big data has been a fast-changing research area with many new opportunities for applications in manufacturing. This paper presents a systematic literature review of the state-of-the-art of big data in manufacturing. Six key drivers of big data applications in manufacturing have been identified. The key drivers are system integration, data, prediction, sustainability, resource sharing and hardware. Based on the requirements of manufacturing, nine essential components of big data ecosystem are captured. They are data ingestion, storage, computing, analytics, visualization, management, workflow, infrastructure and security. Several research domains are identified that are driven by available capabilities of big data ecosystem. Five future directions of big data applications in manufacturing are presented from modelling and simulation to real-time big data analytics and cybersecurity.     

BDMasker: Dynamic Data Protection System for Open Big Data Environment

Big data has become a national basic strategic resource, and the opening and sharing of data is the core of China''s big data strategy. Cloud native technology and lake-house architecture are reconstructing the big data infrastructure and promoting data sharing and value dissemination. The development of the big data industry and technology requires stronger data security and data sharing capabilities. However, data security in an open environment has become a bottleneck, which restricts the development and utilization of big data technology. The issues of data security and privacy protection have become increasingly prominent both in the open source big data ecosystem and the commercial big data system. Dynamic data protection system under the open big data environment is now facing challenges in regards such as data availability, processing efficiency, and system scalability. This paper proposes the dynamic data protection system BDMasker for the open big data environment. Through a precise query analysis and query rewriting technology based on the query dependency model, it can accurately perceive but does not change the original business request, which indicates that the whole process of dynamic masking has zero impact on the business. Furthermore, its multi-engine-oriented unified security strategy framework realizes the vertical expansion of dynamic data protection capabilities and the horizontal expansion among multiple computing engines. The distributed computing capability of the big data execution engine can be used to improve the data protection processing performance of the system. The experimental results show that the precise SQL analysis and rewriting technology proposed by BDMasker is effective. The system has good scalability and performance, and the overall performance fluctuates within 3% in the TPC-DS and YCSB benchmark tests.