一种新的多类支持向量机决策方法及其应用

讨论和比较了现有的几种多类SVM方法．在此基础上,提出了一种组合多个两类分类器结果的多类SVM决策方法．在该方法中,定义了新的决策函数,其值是在传统投票决策值的基础上乘以不同分类器的权重．新的多类SVM在一定程度上解决了传统投票决策方法的不可分区域问题,因此具有更好的分类性能．最后,将新方法作为关键技术应用于故障诊断实例,实际诊断结果证明了所提多类SVM决策方法的优越性．     

基于投票模型的元搜索排序合成算法

排序合成问题是元搜索引擎研究的一个重要方面。该文分析了基于投票模型的排序合成问题。在讨论2个常用的投票规则Borda和Condorcet的基础上，介绍了用图论算法实现的淘汰投票算法，包括Kemeny算法。针对Kemeny算法是NP-hard问题，提出了一种易于实现的启发式淘汰投票算法，并且利用TREC数据集进行实验比较这些方法。实验结果表明，淘汰投票算法与Borda算法执行效果相当，有时甚至超过Borda算法。     

有限理性条件下投票过程的元胞自动机模型研究

有限理性理论认为个体的决策能力是有限的,本文建立元胞自动机（CA）模型研究群体决策中有限理性个体的投票过程,给出了CA的演化规则,模拟了相互邻接的个体问的相互作用,观测到了投票过程的一些微观现象。元胞自动机理论将是推动自然科学和社会科学发展的一个有效工具。     

决策表的一种知识约简与规则获取方法

粗糙集理论是一种新型的数据挖掘和决策分析方法,利用粗糙集理论进行决策表的知识约简与决策规则挖掘已经成为研究热点。文中介绍了粗糙集的基本理论,在此基础上运用该理论对从决策表中获取最小规则进行了研究,提出了决策表约简的启发式方法,并通过一个具体实例详细说明了决策规则获取过程,实例分析表明了其有效性。     

试析量子投票模式

对量子投票中的移动式投票和分布式投票进行研究,分别给出了协议的一般性描述和安全性分析.研究表明,在量子资源上,移动式投票的量子态制备较为容易,而在安全性方面,分布式投票由于对某些攻击是免疫的,因此更为安全.     

决策表的一种知识约简与规则获取方法

粗糙集理论是一种新型的数据挖掘和决策分析方法，利用粗糙集理论进行决策表的知识约简与决策规则挖掘已经成为研究热点。文中介绍了粗糙集的基本理论，在此基础上运用该理论对从决策表中获取最小规则进行了研究，提出了决策表约简的启发式方法，并通过一个具体实例详细说明了决策规则获取过程，实例分析表明了其有效性。     

基于多分类器组合的一种手写体数字识别方法

提出了一种手写体数字识别系统.该系统由三级分类器组成第一级提取交叉点、闭和圆等结构特征,并用模板匹配的方法进行分类;第二级由两个并行的神经网络分类器组成,每个分类器分别使用不同的统计特征;第三级是综合分类器,它将第二级的输出作为输入,根据投票规则得到最后的输出结果.多分类器组合可以集合分类器的优点,提高整个识别系统的识别精度和可靠性.     

基于粗糙集理论的票据打印字符识别系统

票据打印字符识别有着广泛的应用前景.但南于票据打印字符本身所具有多噪声、字体及字符大小多变等特点,导致了票据打印字符识别始终是一项技术难题.提出了一种基丁粗糙集理论的票据打印字符识别系统,首先采集打印数字字符构成训练样奉集,提取训练样本的特征向量,建立样本集的特征矩阵.然后基于粗糙集理论,采用一种有效的属性集约简方法和一种新的规则提取方法,从中导出三个规则集合,基于该三个规则集对测试样本分别进行决策,最后对决策结果进行决策级融合.将该方案应用于纸制飞机票上打印数字字符的识别上,实验证明了该方法的可行性及有效性.     

投票理论研究现状及其展望

社会选择理论是研究如何表达和聚合个体选择的一门学问.而社会选择理论与计算机科学的融合产生了称为计算社会选择的交叉学科,该学科成为社会计算的重要研究内容之一,在人工智能、经济和计算性理论领域引起了轰动.其一方面引入了复杂性分析和算法设计等计算机学科中常用的技术来对社会选择机制进行研究;另一方面也通过引入社会选择理论中的概念来推动计算机技术的发展,特别是在多智能体系统研究中有着成功的应用.投票理论是计算社会选择中最重要的研究主题之一.首先介绍常见的投票方法以及投票理论的形式化框架;再对投票理论中所关心的操纵问题做分析;然后介绍在组合域上的投票;最后对其他相关问题作简要介绍,并对该领域未来的发展与应用做出展望.     

基于密度聚类和投票判别的三维数据去噪方法

本文介绍的是一种有效发现和去除三维数据噪音方法,它既能改善去除噪音的效果,又能保持特征信息。该方法是一种两阶段噪音数据处理方法。该方法首先通过密度聚类将数据分类为正常簇集合、疑似簇集合、异常簇集合,然后利用正常簇集合中的点对疑似簇集合中各点进行投票判断,最终得到一个合理的三维点云数据模型。实验结果证明,本方法能够有效去除制造类工件模型的三维点云数据中的噪音数据,同时能良好保持模型表面的特征,加快处理效率。     

Linking analytic hierarchy process and social choice methods to support group decision-making in water management

The social choice (SC) theory is in close relation with multicriteria decision-making (MCDM), especially in group decision contexts. SC theory includes various voting systems while MCDM is represented by utility and outranking methods; among utility models, the analytic hierarchy process (AHP) is probably the most popular in group decision support. In this paper, we investigate two possible contexts in modeling decentralized decision problems in water management. The first is based on AHP only and two group aggregation techniques. The second one assumes the AHP application in subgroups, while at a group level, aggregation is performed by the SC voting procedures. Comparative analyses show good agreement of the results when two methodologies are applied as the decision support to the water committee of the San Francisco river basin in Brazil. The second methodology (called AHP + SC) is considered more promising for implementation in real-decision situations in water management.     

Committees of Systems of Linear Inequalities

The conceptual issues of the theory of committee decision rules were considered, and its close relationship with the theory of substantiation of collective decision making and teaching neural networks was demonstrated. The problem of the minimum committee of an incompatible system of constraints, which arises at the stage of constructing the committee decision rule with a small number of elements, was discussed by itself. The problem of the minimum committee is known to be NP-hard. Results concerning estimation of the computational complexity of allied problems were obtained. To solve the problem of minimum committee of an incompatible system of linear inequalities, an effective approximate algorithm was suggested, its correctness was proved and computational complexity and guaranteed precision were estimated.     

粗糙集在客户分类中的应用研究

粗糙集理论的主要思想是在保持分类能力不变的前提下,通过属性约简和值约简,提取决策规则。设计了一个基于粗糙集的客户分类模型,并利用粗糙集的知识约简和决策规则提取算法对超市客户进行了分析。通过决策表约简,剔除冗余属性、消除过剩规则。最后得出了属性约简的最小化结果以及决策规则。     

The learnability of voting rules

Scoring rules and voting trees are two broad and concisely-representable classes of voting rules; scoring rules award points to alternatives according to their position in the preferences of the voters, while voting trees are iterative procedures that select an alternative based on pairwise comparisons. In this paper, we investigate the PAC-learnability of these classes of rules. We demonstrate that the class of scoring rules, as functions from preferences into alternatives, is efficiently learnable in the PAC model. With respect to voting trees, while in general a learning algorithm would require an exponential number of samples, we show that if the number of leaves is polynomial in the size of the set of alternatives, then a polynomial training set suffices. We apply these results in an emerging theory: automated design of voting rules by learning.     

基于概率图的三支决策模型研究

概率图模型是一类用图形模式表达基于概率关系的模型的总称,用该模型解决损失代价问题已成为当前的研究热点。结合概率图和三支决策理论,提出了基于概率图的三支决策模型。该模型通过对数据进行分析,构造其Bayes网络;并根据模型中节点的相互依赖关系,计算出条件概率分布函数;结合查询变量的先验概率和三支决策损失代价函数,建立了相应的决策规则,给出了概率推理决策中代价最小化问题的一种解决方法。最后通过教学评估实例验证了该模型的有效性。     

Reliability measure assignment to sonar for robust target differentiation

This article addresses the use of evidential reasoning and majority voting in multi-sensor decision making for target differentiation using sonar sensors. Classification of target primitives which constitute the basic building blocks of typical surfaces in uncluttered robot environments has been considered. Multiple sonar sensors placed at geographically different sensing sites make decisions about the target type based on their measurement patterns. Their decisions are combined to reach a group decision through Dempster-Shafer evidential reasoning and majority voting. The sensing nodes view the targets at different ranges and angles so that they have different degrees of reliability. Proper accounting for these different reliabilities has the potential to improve decision making compared to simple uniform treatment of the sensors. Consistency problems arising in majority voting are addressed with a view to achieving high classification performance. This is done by introducing preference ordering among the possible target types and assigning reliability measures (which essentially serve as weights) to each decision-making node based on the target range and azimuth estimates it makes and the belief values it assigns to possible target types. The results bring substantial improvement over evidential reasoning and simple majority voting by reducing the target misclassification rate.     

Set-valued ordered information systems

Set-valued ordered information systems can be classified into two categories: disjunctive and conjunctive systems. Through introducing two new dominance relations to set-valued information systems, we first introduce the conjunctive/disjunctive set-valued ordered information systems, and develop an approach to queuing problems for objects in presence of multiple attributes and criteria. Then, we present a dominance-based rough set approach for these two types of set-valued ordered information systems, which is mainly based on substitution of the indiscernibility relation by a dominance relation. Through the lower/upper approximation of a decision, some certain/possible decision rules from a so-called set-valued ordered decision table can be extracted. Finally, we present attribute reduction (also called criteria reduction in ordered information systems) approaches to these two types of ordered information systems and ordered decision tables, which can be used to simplify a set-valued ordered information system and find decision rules directly from a set-valued ordered decision table. These criteria reduction approaches can eliminate those criteria that are not essential from the viewpoint of the ordering of objects or decision rules.     

粗糙集分类算法中的近似决策规则和规则匹配方法

粗糙集分类算法在应用标准决策规则进行新对象分类时,经常碰到决策规则与新对象不完全匹配的情况。因此,近似决策规则和部分匹配方法常用于提高决策规则与新对象匹配的可能性。本文在概述和比较两种近似决策规则生成算法的基础上,以一个文本分类系统为例,提出了一种综合的、更有效的近似决策规则生成算法。文章还介绍了几种通用的规则匹配方法,提出了一系列实用的完全匹配和部分匹配公式。实验表明,新提出的近似决策规则生成算法和规则匹配公式能够有效地提高决策规则与新对象的匹配可能性与准确性。     

Graded dominance interval-based fuzzy objective information systems

Dominance interval-based fuzzy objective information systems are generalized models of single-valued fuzzy information systems. By introducing a graded dominance relation to dominance interval-valued fuzzy objective information systems, we establish a graded dominance interval-valued rough set model (RSM), which is mainly based on replacing the indiscernibility relation in classical rough set theory with the graded dominance interval-valued relation. Furthermore, in order to simplify knowledge representation and extract nontrivial simpler graded dominance interval fuzzy decision rules, we propose two attribute reduction approaches to eliminate the redundant condition attributes that are not essential from the viewpoint of graded dominance interval-valued fuzzy decision rules. These results are helpful for decision-making analysis in dominance interval-valued fuzzy objective information systems.