灰局势决策方法在决策层融合目标识别中的应用

本文提出了一种基于灰局势决策的决策层融合目标识别算法,并利用各子源传感器判决结果包含的动态信息,通过计算灰关联系数对各类目标进行了加权处理。实验中利用上述方法对五类目标雷达观测数据的分类结果进行了融合,其结果表明该方法的目标识别性能与子源传感器相比得到了有效地提高。     

隐私保持的决策树分类挖掘

 隐私保持的数据挖掘是目前数据挖掘领域的重要研究方向之一,其首要研究内容是开发在不泄露隐私数据的前提下进行数据挖掘的方法.决策树是分类挖掘的一种重要方法,也是目前隐私保持的数据挖掘领域中少有的被深入研究了的分类方法.针对目前尚未对隐私保持的决策树挖掘方法进行系统总结的问题,本文对该领域进行综述.首先对问题背景进行介绍,随后介绍了该领域的研究现状,对现有方法进行了分类和总结,最后总结出该领域进一步研究的方向.     

盾构机掘进过程中的决策支持系统

盾构机掘进过程中的多系统协调控制,对于有效地控制盾构机隧道施工引起的地表变形、保证盾构安全施工和降低隧道掘进能源消耗都是极其重要的。提出了基于现场观测数据的盾构机掘进决策支持系统模型。根据盾构机掘进过程中观测的刀盘贯入度、刀盘转速、刀盘扭矩、盾构机推力和推进速度等观测数据,在地层模糊聚类分析的基础上,实现随机分布、复杂性地层特征在线辨识。以盾构机掘进参数现场观测数据为基础,利用神经网络具有的非线性映射能力、自组织和自适应能力,实现土仓压力模型的自适应建模,解决非线性、时变性、随机性和时滞性的土仓压力分布模型建模与预测控制问题。通过实时调节和控制不同位置推进油缸的推力,减少由于复杂地层的不均匀性所带来的盾构机姿态的偏差,提高隧道的成型精度。针对不同的地层,优化确定与其相适应的盾构机掘进参数,降低掘进能耗与刀具磨损,提高盾构机控制系统的顺应性。     

On-Demand Data Broadcasting for Mobile Decision Making

The wide spread of mobile computing devices is transforming the newly emerged e-business world into a mobile e-business one, a world in which hand-held computers are the user's front-ends to access enterprise data. For good mobile decision making, users need to count on up-to-date, business-critical data. Such data are typically in the form of summarized information tailored to suit the user's analysis interests. In this paper, we are addressing the issue of time and energy efficient delivery of summary tables to mobile users with hand-held computers equipped with OLAP (On-Line Analytical Processing) front-end tools. Towards this, we propose a new on-demand scheduling algorithm, called STOBS, that exploits the derivation semantics among OLAP summary tables. It maximizes the aggregated data sharing between mobile users and reduces the broadcast length for satisfying a set of requests compared to the already existing techniques. The algorithm effectiveness with respect to access time and energy consumption is evaluated using simulation.     

Decision feedback equalization

As real world communication channels are stressed with higher data rates, intersymbol interference (ISI) becomes a dominant limiting factor. One way to combat this effect that has recently received considerable attention is the use of a decision feedback equalizer (DFE) in the receiver. The action of the DFE is to feed back a weighted sum of past decision to cancel the ISI they cause in the present signaling interval. This paper summarizes the work in this area beginning with the linear equalizer. Three performance criteria have been used to derive optimum systems; 1) minimize the noise variance under a "zero forcing" (ZF) constraint i.e., insist that all intersymbol interference is cancelled, 2) minimize the mean-square error (MMSE) between the true sample and the observed signal just prior to the decision threshold, and 3) minimize the probability of error (Min Pe). The transmitter can be fixed and the receiver optimized or one can obtain the joint optimum transmitter and receiver. The number of past decisions used in the feedback equalization can be finite or infinite. The infinite case is easier to handle analytically. In addition to reviewing the work done in the area, we show that the linear equalizer is in fact a portion of the DFE receiver and that the processing done by the DFE is exactly equivalent to the general problem of linear prediction. Other similarities in the various system structures are also shown. The effect of error propagation due to incorrect decisions is discussed, and the coaxial cable channel is used as an example to demonstrate the improvement available using DFE.     

数据仓库技术在决策支持系统中的应用

介绍了数据仓库和决策支持系统的定义及相关技术,并论述了数据仓库技术应用在决策支持系统中的方案及其开发方法。     

Decision level ensemble method for classifying multi-media data

Wireless Networks - In the digital era, the data, for a given analytical task, can be collected in different formats, such as text, images and audio etc. The data with multiple formats are called...     

分类判别的覆盖算法研究

基于数据的机器学习是研究从观测数据出发寻找规律,并利用这些规律对未来数据进行预测.该文提出一种新的分类判别方法--覆盖算法,其主要过程是利用某种覆盖规则算法寻找一些训练样本集的支撑点(代表点),在决策的时候仅需计算待分类样本与支撑覆盖点之间的距离并进行比较,与之最近的支撑点所在类别即为代分类样本的类别.而支撑点仅占全部训练样本的一部分,所以相比最近邻方法具有较小运算量和存储量的优点.另一方面,覆盖算法主要是样本之间的距离运算,不需要像SVM那样考虑核函数的选择问题,因此更适用于大数据量的自动分类问题.对正常星系和恒星两类光谱数据进行实验,结果表明,覆盖算法具有较好的鲁棒性、较高的分类正确率.     

防空火箭烟幕弹干扰决策

介绍了防空火箭烟幕弹的外弹道特点,比较分析了获取外弹道数据三种方法的优缺点,给出了外弹道拟合公式的建立过程．在此基础上,提出了两种防空火箭烟幕弹干扰辅助决策算法．最后,对防空火箭烟幕弹干扰的弹道风修正算法进行了探讨．     

Dominance Lagrange Optimized Rule Generation for Decision Table Evaluation

Mobile Networks and Applications - Manipulating the volumes of data and recognizing significant information was the challenging task in organizing and managing with uncertain data. In data...     

决策支持系统—企业MIS的最终实现目标

本文针对信息技术及高科技行业提出了通过数据集市和数据仓库的建设，实现企业决策支持系统的概念。并期望通过本文为建设企业管理信息系统提供了一个可行的解决方法。     

基于OLAP的移动决策支持系统

国内电信运营商间的竞争越来越激烈,单纯的价格战将对竞争双方造成损失。因此,电信运营商都在寻求改善服务质量、提高市场竞争力的方法。本文研究了数据仓库数据析取与联机分析处理的有效方法,并采用OLAP技术建立了以“客户行为分析”为主题的移动决策支持系统,使得用户能方便、快速地从多个角度观察、分析数据,同时为数据挖掘应用提供了良好的基础。     

基于决策树分类算法异构数据的索引优化

海量数据的索引是提高分布式环境下海量数据的查询重要手段。为了构建高效的索引结构,人们提出了多种异构数据索引优化方法。文中给出了基于决策树分类算法下的索引优化方法。基于决策树分类算法构建索引决策树,利用该索引决策树对各个子空间表的属性列进行决策,建立索引表,根据索引表数据建立索引,再根据各子空间上的索引构建全局索引。该二级索引结构为快速定位索引信息提供了技术支持。实验结果表明,索引决策树是一个对优化异构数据索引合适的方法。     

基于决策树算法的信息资产识别

资产识别是网络安全动态风险评估过程的首要环节,在基于层面的信息资产划分方法上,通过数据挖掘技术中的决策树算法,研究了信息资产识别问题。实现了数据挖掘技术在资产识别中的应用。在对C4．5算法的基本流程进行描述的基础上,以信息资产数据集中的操作系统为例,利用C4．5算法实现了对数据的正确分类,并进行了规则的提取,为以后进行资产识别提供了依据,并指明了今后该算法在信息资产识别其他方面的研究方向。     

装备维修智能决策支持系统研究

从系统的功能需求入手介绍传统的智能决策支持系统的结构模型,分析装备维修智能决策支持系统的功能实现,按照智能维修的目标设计了系统的总体构架。数据仓库的建立和分析采用传统的方法;对装备维修资料进行分析组建相应的知识库;分析建立模型条件结合知识库建立模型库;用数据挖掘方法处理数据仓库、知识库和模型库中的知识实现系统的智能决策。结果证明该方法在装备维修智能决策支持系统中得到很好的应用。     

Decision feedback-based demodulation of doubly differential PSKsignals

A decision feedback-based differential demodulator of double differential phase-shift keying (DDPSK) signals is presented, It is comprised of two consecutively connected differential detectors, which are based on certain reference signals obtained from previously detected data and received signals. Computer simulations demonstrate that the bit error rate performance of the proposed demodulator is comparable to that of the coherent receiver of DDPSK signals     

Decision feedback equaliser based on fuzzy logic

A method for reducing the effect of error propagation in a decision feedback equaliser (DFE) by using fuzzy logic is presented. A simple membership function is proposed and adaptively determined from data. Simulation results are presented to show that the combination of the DFE and fuzzy logic is promising     

自适应模糊决策树算法在数据流挖掘中的应用

在网络的许多应用中数据是以流的形式存在的,例如网络流、传感器数据,以及网页点击流等,分析和挖掘这类数据,可以发现某中有价值的信息。在此,针对数据流挖掘算法中出现的一些问题（如概念漂移问题）,提出了一种自适应模糊决策树的优化算法。该算法对于解决处理数据流概念中的漂移问题有较好的效果。     

A Decision Tree Approach for Predicting Smokers’Quit Intentions

This paper presents a decision tree approach for predicting smokers’ quit intentions using the data from the International Tobacco Control Four Country Survey. Three rule-based classification models are generated from three data sets using attributes in relation to demographics, warning labels, and smokers’ beliefs. Both demographic attributes and warning label attributes are important in predicting smokers’ quit intentions. The model’s ability to predict smokers’ quit intentions is enhanced, if the attribu...     

智能化决策支持系统框架的研究与设计

主要在对智能化决策支持系统（IDSS）进行研究并分析目前研究成果的基础上,根据IDSS的开发过程和特点,融合Web开发的MVC设计模式的思想,采用专家系统（ES）、数据仓库（DW）、联机分析处理（OLAP）和数据挖掘（DM）等技术,提出一种综合IDSS的结构框架,将其划分为4个层次,并对各个层次的作用分别进行说明。