Averaging over decision trees

Pruning a decision tree is considered by some researchers to be the most important part of tree building in noisy domains. While there are many approaches to pruning, the alternative of averaging over decision trees has not received as much attention. The basic idea of tree averaging is to produce a weighted sum of decisions. We consider the set of trees used for the averaging process, and how weights should be assigned to each tree in this set. We define the concept of afanned set for a tree, and examine how the Minimum Message Length paradigm of learning may be used to average over decision trees. We perform an empirical evaluation of two averaging approaches, and a Minimum Message Length approach.This work has been carried out with the support of the Defence Research Agency, Malvern.     

DECISION MAKING - THE ANALYTIC HIERARCHY AND NETWORK PROCESSES (AHP/ANP)

This is the first part of an introduction to multicriteria decision making using the Analytic Hierarchy Process (AHP) and its generalization, the Analytic Network Process (ANP). The discussion involves individual and group decisions both with the independence of the criteria from the alternatives as in the AHP and also with dependence and feedback in the entire decision structure as in the ANP. This part explains the Analytic Hierarchy Process, with examples, and presents in some detail the mathematical foundations. An exposition of the Analytic Network Process and its applications will appear in later issues of this journal.     

基于属性约简的决策表算法

Rough Set理论通过属性约简和决策规则约简,达到发掘知识并简化知识的目的。本文首先给出了属性重要性的度量方法,然后对条件属性的简化进行了阐述,最后,讨论了一致决策表最小算法的三类优化问题。并通过一个实例证明了这种算法。     

The futility of decision making research

We have each spent more than 50 years doing research that has had little impact. Even more lamentable is that our field, judgment and decision making (JDM), has on the whole had little impact during that span. We attribute that failure to the use of methodologies that emphasize testing models rather than looking for differences in behavior. The “cognitive revolution” led the field astray, toward the goal of studying model fit rather than comparing observable results. With modeling as the goal, experimentation was stultified. Simple tasks became dominant. Although a poor metaphor for real decision making, the gambling paradigm has lasted forever because the inputs to the decision are known to the researcher and thus easily modeled.     

状态表法及其应用案例分析

本文首先介绍了状态表的基本概念,然后根据一个应用系统原型,阐述了创建状态表的方法,以及如何利用状态表对系统进行控制,使其在某种随机因素的扰动下,能自动工作于系统最佳工作状态。     

基于决策树的轨道电路故障诊断知识表示方法研究

针对ZPW-2000A无绝缘轨道电路故障逻辑机理模糊的问题,本文采用了基于决策树的轨道电路专家系统知识表示方法.该方法首先将轨道电路故障影响较大的特征向量样本建立故障决策表,然后运用最小信息熵算法对属性值离散化,根据决策树算法快速学习及分类的特点对离散后的数据样本进行训练学习,生成故障决策树后进行知识规则的获取,在专家系统的知识库中以产生规则存储.通过对ZPW-2000A无绝缘轨道电路的实例分析验证了该方法在轨道电路专家系统知识表示与获取中的有效性和实用性.     

基于Trio控制器的吸收式造波机运动控制系统设计

设计了基于Trio控制器的吸收式造波机运动控制系统,确定了其PC和运动控制器相结合的控制方式.选择了Trio控制器,并结合EtherNet/IP和SERCOS通信协议实现数据交换.上位机程序设计采用C语言和Fortran混合编程方式,对执行机构的控制由Trio控制器自带Trio Basic语言编写运动控制程序程实现.实验表明,设计的吸收式造波机控制系统得到的波面最稳定,能获得期望波高.     

C&RT决策树在民办高校招生中的应用分析

以往的数据挖掘技术的应用大多是在金融领域,而在其他领域里面应用不是很多,如在高校招生中的应用更是如此。在数据挖掘中,决策树提供了一种展示类似在什么条件下会得到什么值这类规则的方法,是用二叉树形图来表示处理逻辑的一种技术。特别适合于判断因素比较少、逻辑组合关系不复杂的情况。它可以直观、清晰地表达加工的逻辑要求,满足用户需求,为管理决策者提供相应的决策支持。以某高校招生数据作为基础信息为依据,对高校招生中生源地与报到情况等进行CART(Classification and Regression Tree)分析,将对决策树分析的应用做进一步的阐述。     

基于动态规划的企业投资决策模型

为了保证企业投资决策最优的投资效果,企业应把投资决策过程分为多个阶段.建立了以获得利润最大化,同时把投资总风险控制在可承受范围内为目标的基本动态规划模型.该模型把一个多阶段的投资问题转化为多个单阶段的问题,从而求解整个投资阶段的最优决策问题就转化成求解一系列单个投资阶段中的最优问题.     

基于两种信息融合的turbo码帧同步器研究

提出了一种基于两种信息融合的turbo码帧同步器。对turbo码1步迭代外信息统计发现,帧同步与帧失步时turbo码的外信息的统计特性存在明显的差异。在信噪比极低的情况下,传统的基于同步码相关的方法已不能满足turbo码强译码能力的要求。为了进一步降低帧同步差错率,同时又不给系统带来太大计算延迟,对turbo译码器1次迭代后的最大后验信息和同步码字相关后的最大似然信息融合成一个置信度量,仿真表明：只需很小的额外计算量,这种方法能够获得比传统方法更低的帧同步差错率。