大数据关联关系度量研究综述

大数据关联性分析是大数据挖掘的基础,一个好的关联性度量是实施关联分析的关键。本文首先指出大数据时代关联度量面临的挑战和研究现状,从关联关系度量的构造角度出发,对现有的关联关系度量进行整理,归纳总结了这些关联关系的性质和适用条件。在回顾关联度量发展历程的基础上,结合大数据时代关联关系的特点,提出构造关联度量可能满足的条件。最后针对多模态数据关联关系度量的若干问题进行探讨和梳理,从3个角度出发,提出应对多模态数据空间转换的挑战,以引起对该领域更深入的思考与研究工作,从而促进大数据挖掘工作的进展。     

Study of a Class of Almost‐controllable Discrete‐time Bilinear Systems

This paper studies the controllability of a class of time‐invariant discrete‐time bilinear systems. Although the system is not controllable in the whole space, there is a very large region where control is effective. Results show that the uncontrollable region of this kind of bilinear system has a Lebesgue measure of zero. In other words, for almost any initial state and any terminal state in the state space, the former can be transferred to the latter. Further, a necessary condition for near controllability is presented. Therefore, the results in this paper unify and generalize the corresponding conclusions in the literature. ©2014 Chinese Automatic Control Society and Wiley Publishing Asia Pty Ltd     

Complete controllability for abstract measure differential systems

In this paper, we investigate the complete controllability for abstract measure differential systems. Firstly, we introduce several new concepts about complete controllability for abstract measure differential systems. Then, on the basis of the Sadovskii fixed‐point theorem, we give sufficient conditions for complete controllability for a class of abstract measure differential systems. The compactness of the semigroup generated by some operator is unnecessary in this paper, and we show that our results, dealing with complete controllability problem for an ordinary differential system in infinite‐dimensional Banach space, are also less conservative than that in the previous literature. Copyright © 2012 John Wiley & Sons, Ltd.     

Stability Analysis for Delayed Neural Networks with Discontinuous Neuron Activations

This paper considers a class of delayed neural networks with discontinuous neuron activations. Based on the theory of differential equations with discontinuous right‐hand sides, some novel sufficient conditions are derived that ensure the existence and global exponential stability of the equilibrium point. Moreover, by adopting the concept of convergence in measure, convergence behavior for the output is discussed. The obtained results are independent of the delay parameter and can be thought of as a generalization of previous results established for delayed neural networks with Lipschtz continuous neuron activations to the discontinuous case. Finally, we give a numerical example to illustrate the effectiveness and novelty of our results by comparing our results with those in the early literature.     

基于Value双向近似推理的系统决策方法

先提出Vague值间的一种新的相似度,并基于有序加权聚类(OWA)算子给出Vague集间的新的加权相似测度,其中的权重是由决策者根据实际决策需求来确定的,具有一定的弹性;另外通过Vague输入事实与Vague决策规则的前件(或后件)之间的相似匹配度,得到一个简单可行的双向近似推理方法,这可为复杂环境下的系统决策(如工业检测和医疗诊断等领域问题)提供有效的支持工具。     

系统L在非均匀概率空间下命题的真度理论

将经典二值命题逻辑L中公式的真度概念推广到势为2的非均匀概率空间上;当p∈(0,1)时,证明了全体公式的真度值之集在[0,1]中没有孤立点;利用真度定义公式间的p-相似度和伪距离,进而定义了p-逻辑度量空间,证明了该空间没有孤立点,并在此空间中提出了三种不同类型的近似推理模式。     

实时视频中的车辆运动轨迹的提取和聚类

利用运动物体轨迹的方向性、运动性和相互关系等典型特征对物体的运动轨迹进行聚类。首先利用改进的加权矢量Hausdorff距离作为度量运动物体轨迹相似度的方法,从而使之适用于空间距离有差别的运动物体轨迹的谱聚类问题;然后基于等周分割（ISO）算法,构造轨迹相似度矩阵,完成轨迹的粗聚类;最后利用轨迹的方向性特征和轨迹类间距对轨迹进行二次聚类,得到最终的轨迹聚类结果。     

软件可靠性测试用例的复用度量

针对目前基于复用的软件可靠性测试缺少对测试用例复用度量的现实问题,在构建测试用例复用模型的基础上,首先应用特征分析法求解测试用例各复用资源的权重,然后再分别计算它们在测试中的复用度,最后求解得到测试用例的平均复用度。从定量的角度解决了测试用例的复用度量问题。     

一种基于矩阵度量的缺陷管理流程的改进与实践

本文在分析软件过程中缺陷类型、缺陷注入、缺陷识别的基础上,对传统缺陷管理流程进行改进,增加了缺陷排除有效性的度量方法;然后提出一种实用的软件缺陷管理流程,建立了一个以软件缺陷生命周期为基础的度量模型,并给出了相应的缺陷矩阵度量方法;最后把该缺陷管理流程和度量方法应用在某公司的两个软件项目中,对各阶段的缺陷进行了度量,经实践和数据分析得出,运用此缺陷管理流程和度量方法可以为开发团队设定具体阶段目标和质量计划提供数据基础,为过程控制、过程评价、持续改进等提供量化管理的基础,表明本文改进后的缺陷管理流程和度量方法模型是有效的。     

基于可能性测度的计算树逻辑

首先,提出了可能的Kripke结构的定义,建立了可能的Kripke结构的可能性测度空间,并分析了可能的Kripke结构的一系列性质,即任一路径转移的可能性可由其初始状态的可能性分布与各转移的可能性取下确界而得到;依据可能的Kripke结构所定义的可能性测度具有其合理性等等。其次,给出了可能性计算树逻辑(PoCTL)的概念,讨论了两个PoCTL状态公式以及PoCTL与经典计算树逻辑(CTL)公式的等价性。最后,证明了PoCTL公式有与CTL*公式中"一致性"相对应的公式。