基于频繁项集发现的匿名隐私保护算法*

通过定义考虑权重的匿名表效用度量函数,用于在泛化步骤决定下一个泛化路径以取得较好的泛化效果,在此基础上提出利用频繁项集发现思想的微观数据表匿名隐私保护算法ABFI(algorithm based on frequent set mining),匿名过程仅仅对不满足隐私保护要求等价组中准码属性取值进行泛化。实验结果表明,该方法可以减少信息损失,求解得到更加符合数据分析任务需求的局部最优匿名表。     

一种考虑属性权重的隐私保护数据发布方法

k-匿名模型是数据发布领域用于对原始待发布数据集进行匿名处理以阻止链接攻击的有效方法之一,但已有的k-匿名及其改进模型没有考虑不同应用领域对匿名发布表数据质量需求不同的问题.在特定应用领域不同准码属性对基于匿名发布表的数据分析任务效用的贡献程度是不同的,若没有根据发布表用途的差异区别处理各准码属性的泛化过程,将会导致泛化后匿名发布表数据效用较差、无法满足具体数据分析任务的需要.在分析不同应用领域数据分析任务特点的基础上,首先通过修正基本ODP目录系统建立适用于特定问题领域的概念泛化结构;然后在泛化过程中为不同准码属性的泛化路径设置权重以反映具体数据分析任务对各准码属性的不同要求;最后设计一种考虑属性权重的数据匿名发布算法WAK(QI weight-aware k-anonymity),这是一种灵活地保持匿名发布表数据效用的隐私保护问题解决方案.示例分析和实验结果表明,利用该方案求解的泛化匿名发布表在达到指定隐私保护目标的同时,能够保持较高的数据效用,满足具体应用领域特定数据分析任务对数据质量的要求.     

面向缺失数据的数据匿名方法

在数据发布过程中,为了防止隐私泄露,需要对数据的准标识符属性进行匿名化,以降低链接攻击风险,实现对数据所有者敏感属性的匿名保护.现有数据匿名方法都建立在数据无缺失的假设基础上,在数据存在缺失的情况下会直接丢弃相关的记录,造成了匿名化前后数据特性不一致.针对缺失数据匿名方法进行研究,基于k-匿名模型提出面向缺失数据的数据匿名方法KAIM(k-anonymity for incomplete mircrodata),在保留包含缺失记录的前提下,使在同一属性上缺失的记录尽量被分配到同一分组参与泛化.该方法将分组泛化前后的信息熵变化作为距离,基于改进的k-member 算法对数据进行聚类分组,最后通过基于泛化层次的局部泛化算法对组内数据进行泛化.实际数据集的大量实验结果表明,KAIM 造成信息缺损仅为现有算法的43.8%,可以最大程度地保障匿名化前后数据特性不变.     

基于聚类的高效(K,L)-匿名隐私保护

为防止发布数据中敏感信息泄露,提出一种基于聚类的匿名保护算法.分析易被忽略的准标识符对敏感属性的影响,利用改进的K-means聚类算法对数据进行敏感属性聚类,使类内数据更相似.考虑等价类内敏感属性的多样性,对待发布表使用(K,L)-匿名算法进行聚类.实验结果表明,与传统K-匿名算法相比,该算法在实现隐私保护的同时,数据信息损失较少,执行时间较短.     

面向表数据发布隐私保护的贪心聚类匿名方法

为了防范隐私泄露,表数据一般需要匿名处理后发布.现有匿名方案较少分类考察准标识属性概化,并缺少同时考虑信息损失量和时间效率的最优化.利用贪心法和聚类划分的思想,提出一种贪心聚类匿名方法：分类概化准标识属性,并分别度量其信息损失,有利于减小并合理评价信息损失.对元组间距离和元组与等价类距离,建立与最小合并概化信息损失值正相关的距离定义,聚类过程始终选取具有最小距离值的元组添加,从而保证信息损失总量趋于最小.按照k值控制逐一聚类,实现等价类均衡划分,减少了距离计算总量,节省了运行时间.实验结果表明,该方法在减少信息损失和运行时间方面是有效的.     

k-匿名模型中准标识符最佳值的求解问题

准标识符值是影响k-匿名表隐私保护程度和数据质量的关键因素。如何在给定各个准标识符属性泛化树的情况下求解准标识符最佳值,对匿名表在满足隐私保护要求的同时达到最高的数据质量具有重要意义。针对这一问题,证明了准标识符最佳值的求解问题是NP-完全问题,提出了准标识符最佳值的近似求解方法,并给出了准标识符最佳值的近似求解算法;最后,对算法进行了正确性证明和时间复杂度分析。     

k-匿名数据中的数据依赖问题研究

k-匿名隐私保护模型在隐私保护过程中会产生大量k-匿名数据.为研究k-匿名数据中的数据依赖问题,提出一种扩展函数依赖,将经典函数依赖中的被决定属性取值相等这个条件进行扩展,使其取值来自于同一个指定集合.应用结果表明,该扩展函数依赖不仅包括经典函数依赖、垂直函数依赖、水平函数依赖、度量函数依赖的特性,而且可以从数据完整性的角度描述k-匿名数据的约束条件及指导k-匿名隐私保护模型中准标识符的选取.     

MA-Datafly:一种支持多属性泛化的k-匿名方法

Datafly算法是数据发布环境下保护数据隐私的一种k-匿名方法,实现k-匿名时只对准标识符属性集中属性值种类最多的属性进行归纳。当准标识符属性集中只有一个属性的取值多样而其他属性取值具有同质性时,该算法可行。实际应用中数据的取值却往往不具有这种特点。针对这个问题,提出一种自底向上的支持多属性归纳k-匿名算法,并对该算法进行实验测试,结果表明该算法能有效降低原始数据的信息损失并能提高匿名化处理效率。     

基于杂度增益与层次聚类的数据匿名方法

数据匿名是发布数据时对隐私信息进行保护的重要手段之一.对数据匿名的基本概念和应用模型进行了介绍,探讨了数据匿名结果应该满足的要求.为了抵制背景知识攻击,提出了一种基于杂度增益与层次聚类的数据匿名方法,该方法以杂度来度量敏感属性随机性,并以概化过程中信息损失最小、杂度增益最大的条件约束来控制聚类的合并过程,可以使数据匿名处理后的数据集在满足k-匿名模型和l-多样模型的同时,使数据概化的信息损失最小且敏感属性的取值均匀化.在实验部分,提出了一种对数据匿名结果进行评估的方法,该方法将匿名结果和原始数据进行对比,并从平均信息损失和平均杂度2个方面来评估数据匿名的质量.实验结果验证了以上方法的有效性.     

大数据模式分解的隐私保护研究

现有的大多数隐私保护技术往往忽略了敏感属性不同取值和准标识符属性之间存在的特殊关联,并且各领域对数据隐私保护的多方面要求,使得发布的匿名数据需要满足复合隐私约束。对近似敏感属性值和复合隐私约束进行分析,提出了基于大数据模式分解和聚类分析的隐私保护算法。给出了聚类敏感属性值保护相似值方法,设置不同权重的敏感属性,保留重要的属性。使用三维不规则结构矩阵的效用矩阵,来获取精度较高的匿名数据,实现匿名数据的模式分解。在真实数据集上的大量实验结果表明,该算法的数据精确率、数据纠错率都有明显提升,近似攻击率降低。     

Fuzzy measure identification for criteria coalitions using linguistic information

Modeling interactions between criteria in multiple criteria decision analysis (MCDA) is a complex task. Such complexity arises when there are visible redundancies and synergies among criteria, which traditional MCDA methods cannot deal with. The Choquet integral is a model that has been conceived to deal with these issues, but an appropriate fuzzy measure must be defined. This article shows how to compute a fuzzy measure for criteria coalitions using linguistic information efficiently. Due to the complexity to identify an adequate fuzzy measure when the criteria set cardinality increases, the proposed model reduces the effort to determine the measure of each criteria combination by focusing on relevant interactions. Then, this fuzzy measure is used on Choquet integral to establish the best alternative in a decision-making problem. Finally, a comparison between the arithmetic mean, the OWA operator and the proposed method is presented.     

Some fuzzy measure modeled multi-criteria formulations

ABSTRACT

We first describe an approach to multi-criteria making which makes use of a fuzzy measure over the set of criteria to model the user expressed relationship between the criteria. Under this approach we use the Choquet integral, guided by this fuzzy measure, to aggregate an alternative’s satisfactions to the individual criteria. Our focus in this paper is to look at the formulation, in terms of fuzzy measures, of some interesting and novel relationships between the criteria. After noting that the OWA aggregation can be modeled using a cardinality measure we look at the measure formulation of a rule-based expression of a collection of OWA aggregations. We then focus on the measure representation of a user provided well-formed formula in the language of propositional logic expressed relationship between the criteria.     

Importance-based multicriteria decision making with interval valued criteria satisfactions

Multiple-criteria decision problems involve selecting an alternative that best satisfies a collection of criteria as quantified by a scalar corresponding to an aggregation of the alternatives satisfaction to the individual criteria. A fundamental issue is the formulation of decision maker's aggregation function based upon the decision maker's perceived relationship between the criteria. Here, we allow the decision maker to express their perceived relationship between the criteria in terms of information about the criteria importances by providing a fuzzy measure over the criteria such that the measure of any subset of criteria is its importance. With the aid of the Choquet integral, we use this fuzzy measure of importances to construct an aggregation function. As the Choquet integral requires an ordering of an alternatives individual criteria satisfactions, special handling is required in the case when criteria satisfactions are interval valued rather then scalar. Here we use the golden rule representative value in the case of interval values.     

Weight computation of criteria in a decision-making problem by knowledge measure with intuitionistic fuzzy set and interval-valued intuitionistic fuzzy set

The aim of this study is to propose an objective method for determining weights of criteria (also called attributes) based on a new measure of intuitionistic fuzzy information, called knowledge measure, in a real-world multi-criteria decision-making problem under intuitionistic fuzzy and interval-valued intuitionistic fuzzy environment. To address this issue, we first analyze the existing entropy measures and show that their use in objective weight determination process may lead us to produce unreliable weights of criteria by citing appropriate examples. Then we analyze important properties of knowledge measure of intuitionistic fuzzy set (IFS) and also define knowledge measure for interval-valued intuitionistic fuzzy set. Then a new method to determine the weights of criteria is developed on the basis of knowledge measure where information about criteria weights is completely unknown and partly known. A real-life example is presented to illustrate the proposed weight determination method and a comparative analysis is carried out to indicate the practicality and effectiveness of knowledge-based weight-generation method under both intuitionistic fuzzy and interval-valued intuitionistic fuzzy environment. Finally, we formulate the axioms for knowledge measure associated with IFSs and we also propose families (classes) of knowledge measures.     

Multicriteria Decision-Making Using Fuzzy Measures

We are interested in the formulation of multicriteria decision functions based on the use of a measure over the space of criteria. Specifically, the relationship between the criteria is expressed using a monotonic set measure. We then use the Choquet integral to construct decision functions based on the measure. We look at a number of different decision functions generated from specific classes of measures.     

线性独立路径覆盖率的软件测试充分性判别方法

软件测试充分性判别准则是决定一个软件系统是否已经被充分测试的停止准则,而充分性判别准则的关键是它的揭错能力。对充分性判别准则进行了形式化描述,并且讨论了充分性判别准则的性质及准则之间的比较方法。为了给保障软件测试充分性提供理论依据,提出了一个软件测试充分性的度量准则。     

Pythagorean fuzzy TOPSIS for multicriteria group decision-making with unknown weight information through entropy measure

In this study, a new technique for order preference by similarity to ideal solution (TOPSIS)-based methodology is proposed to solve multicriteria group decision-making problems within Pythagorean fuzzy environment, where the information about weights of both the decision makers (DMs) and criteria are completely unknown. Initially, generalized distance measure for Pythagorean fuzzy sets (PFSs) is defined and used to initiate a new Pythagorean fuzzy entropy measure for computing weights of the criteria. In the decision-making process, at first, weights of DMs are computed using TOPSIS through the geometric distance model. Then, weights of the criteria are determined using the entropy weight model through the newly defined entropy measure for PFSs. Based on the evaluated criteria weights, TOPSIS is further applied to obtain the score value of alternatives corresponding to each decision matrix. Finally, the score values of the alternatives are aggregated with the calculated DMs’ weights to obtain the final ranking of the alternatives to avoid the loss of information, unlike other existing methods. Several numerical examples are considered, solved, and compared with the existing methods.     

Objective functions for optimal building design

In this paper, the authors describe 13 investor motivations for initiating a project. Quantitative criteria to measure success in achieving these motivations are listed. One of the these criteria, net present value, is selected as most capable of reflecting key investor motivations. This measure is manipulated to yield a life-cycle costing relationship which can be used as an objective function in the design decision-making process. It is shown that the life-cycle cost equation in its most general form embodies within it a family of commonly used objective functions which describe financial, technical performance and user criteria. Results from a case study in which different objective functions were used are presented.     

云模型与VIKOR集成的多属性群决策方法

针对以往定性概念量化过程中只考虑到模糊性这一不确定性因素的不足,采用云模型从模糊性和随机性两个方面实现语义评价变量的量化,并对专家组的语义评价信息进行集结。针对以往云差异性度量方法的不足,基于云模型构成的本质特点,从云滴分布的角度提出了云距离测度算法,进而提出了云相似度算法。考虑到决策指标间相互影响关系,将云模型与决策试验与实验评估法（Decision-Making Trial and Evaluation Laboratory,DEMATEL）相结合,采用云DEMATEL法对专家主观评价给出的指标初始重要度进行修正,进而计算得到指标权重。采用云距离算法计算备选方案与正、负理想解间的距离,并最终由云VIKOR求得备选方案的妥协解。最后以实例验证了所提方法的可行性和有效性。     

基于模糊积分的潜艇作战能力评估方法

在方案论证阶段用于评估潜艇作战能力的方法,一般都假设各个指标是独立的,而在许多情况下独立性假设是不能成立的。针对系统的方案不确定性,引入了模糊测度的概念,能够对系统指标以及指标之间的关联性,作战能力的影响进行建模,并且给出了一种应用专家判断和作战仿真相结合的模糊测度计算方法。在得到评价准则集的模糊测度以后,引入Cho-quet模糊积分,对潜艇的作战能力进行聚合。最后给出了一个实例验证了方法的有效性。