关联规则的一种定量分析方法

通过引入“结构化规则涵盖集”的概念，运用定量方法来评价对实际工作有用的规则。这样用户不再面对大量繁杂的关联规则，就能直接获取有效指导实践的规则集合。     

A LFP-tree based method for association rules mining in telecommunication alarm correlation analysis

The mining of association rules is one of the primary methods used in telecommunication alarm correlation analysis,of which the alarm databases are very large.The efficiency of the algorithms plays an important role in tackling with large datasets. The classical frequent pattern growth(FP-growth) algorithm can produce a large number of conditional pattern trees which made it difficult to mine association rules in are telecommunication environment.In this paper,an algorithm based on layered frequent pattern tree(LFP-tree) is proposed for mining frequent patterns. Efficiency of this alagorithm is achieved with following techniques:1) All the frequent patterns are condensed into a layered structure,which can save memory time but also be very useful for updating the alarm databases.2) Each alarm item can be viewed as a triple,in which t is a Boolean vaviable that shows the item frequent or not.3) Deleting infrequent items with dynamic pruning can avoid produce conditional pattern sets. Simulation and analysis of algorithm show that it is a valid method with better time and space efficiency,which is adapted to mine association rules in telecommunication alarm correlation analysis.     

Association rule discovery with the train and test approach for heart disease prediction.

Association rules represent a promising technique to improve heart disease prediction. Unfortunately, when association rules are applied on a medical data set, they produce an extremely large number of rules. Most of such rules are medically irrelevant and the time required to find them can be impractical. A more important issue is that, in general, association rules are mined on the entire data set without validation on an independent sample. To solve these limitations, we introduce an algorithm that uses search constraints to reduce the number of rules, searches for association rules on a training set, and finally validates them on an independent test set. The medical significance of discovered rules is evaluated with support, confidence, and lift. Association rules are applied on a real data set containing medical records of patients with heart disease. In medical terms, association rules relate heart perfusion measurements and risk factors to the degree of disease in four specific arteries. Search constraints and test set validation significantly reduce the number of association rules and produce a set of rules with high predictive accuracy. We exhibit important rules with high confidence, high lift, or both, that remain valid on the test set on several runs. These rules represent valuable medical knowledge.     

基于攻击图的多源告警关联分析方法

现有基于攻击图(attack graph)的告警关联分析方法难以全面处理告警关联关系,同时,漏报推断和告警预测带来大量冗余路径误报。针对以上问题提出了基于攻击图的多源告警关联分析算法,能够综合应用图关系和阈值限制进行联动推断和预测,达到更为全面解决攻击图中的告警漏报和减少误报数量的目的。同时,将告警处理算法并行化,提出了AG-PAP告警并行处理引擎。实验表明,该方法能够提升关联分析的有效性和性能表现。     

Rule-based adaptive configuration selection for analog design systems

A configuration selection tool for rule-based analog IC design is presented. A fast determination of proper analog circuit configuration plays an eminent role in the initial design phase. This appears to be quite demanding due to the large number of parameters and side issues involved. In order to ease concept engineering in analog design an adaptive algorithm has been developed for configuration selection. It can be used in design systems that rely on use of expert knowledge stored as a set of rules in a knowledge base, e.g., expert systems, design assistants, etc. The developed learning algorithm also helps eliminate inconsistent, incorrect, and redundant rules in rule bases used in analog design systems. The rule base is continuously adapted, depending on the correctness of the decisions the rules have invoked.     

基于序列模式挖掘的网管告警系统

告警关联分析是网络故障管理中的一个难点,传统方法由于需要引入大量的先验知识而难以适应网络复杂多变的情况.序列模式挖掘作为一种时序数据分析的有效手段,能够自动从告警中提取出有助于关联分析的情景规则.本文首先介绍了与情景规则有关的一些基本概念,然后讨论了挖掘情景规则的常用算法,进而提出了基于序列模式挖掘的告警关联分析的网管系统模型,并实现了一个简单的原型系统,实验证明该系统能够从海量告警中提取出有助于关联分析的情景规则.     

基于候选项集个数上阶的增量式关联规则更新算法

提出了一种有效的增量式关联规则挖掘算法IAR,算法的特点在于:提出并采用了基于候选项集个数上阶的选择扫描数据库的机制,可有效减少数据库的扫描次数;算法是一种通用的增量式算法,提出了最小支持度和数据库均改变时,增量式挖掘中的重要性质,从而可充分利用上一次挖掘的结果,有效减少候选项集的数目.并且提出了基于组合数学和项集等价类理论的计算候选项集个数的上阶的方法.通过大量的数据实验,表明算法的效率比已有的算法有了很大提高.     

Synthesis of Sequential Circuits by Redundancy Removal and Retiming

In this paper we propose a method for synthesizing sequentialcircuits to reduce the number of gates and flip-flops by removingboth combinationally and sequentially redundant faults. In order toremove sequentially redundant faults these faults are converted intocombinationally redundant faults by using retiming techniques and thecombinationally redundant faults can be removed by using a testpattern generation method for combinational circuits. To simplify agiven circuit retiming is utilized for two purposes in thismethod. One is to find sequentially redundant faults and another is toreduce the number of flip-flops and gates. Before and after eachretiming the combinationally redundant faults are removed.Experimental results for ISCAS 89 benchmark circuits show that thismethod can remove many of sequentially redundant faults and canreduce a large number of gates and flip-flops.     

C程序隐式规则自动提取与反例检测

提出一种通用且高效的隐式规则自动提取与反例检测方法,使用频繁闭合项集挖掘技术挖掘包含多种程序元素的编程模式,然后由编程模式产生编程规则;引入正序规则的概念,以避免从同一个编程模式中产生多个冗余规则.在此基础上,提出一种高效的反例检测算法检测违反规则的程序片段.实验结果表明,该方法能够自动提取程序中存在的隐式编程规则,并快速有效地检测违反规则的反例.     

Characterizing network traffic behaviour using granule‐based association rule mining

Association rule mining is one important technique to characterize the behaviour of network traffic. However, mining association rules from network traffic data still have three obstacles such as efficiency, huge number of results and insufficiency to represent the behaviour of network traffic. Aiming to tackle these issues, this paper presents a granule‐based association rule mining approach, called association hierarchy mining. The proposed approach adopts top‐down rule mining strategy to directly generate interesting rules according to subjectively specified rule template hierarchies, which improves the efficiency of rule generation and subjectively filters user uninterested rules. The approach also proposes to prune a new type of redundant rules defined by this research to reduce the number of rules. Finally, the approach introduces the concept of diversity, aiming to select the interesting rules for better interpreting the behaviour of network traffic. The experiments performed on the MAWI network traffic traces show the efficiency and effectiveness of the proposed approach. Copyright © 2016 John Wiley & Sons, Ltd.     

Candidate region correlation for video action detection

The rapid development of deep learning has prompted the development of video action detection technology. However, the accuracy of current video action detection algorithms can be improved further. Previous work has improved feature extraction by optimizing the network structure. In addition, the features of the candidate regions have been optimized by changing the representation of the regions. Although these methods have achieved promising results, they fail to consider the correlation among different candidate regions, generating uninformative (even redundant) candidate regions, and thus usually decrease the detection performance in practice. To address this problem, in this paper we propose a self-attention mechanism for candidate regions, which can help pursue the most informative regions. We obtain the region correlation by simultaneously determining the spatial and temporal correlation among different candidate regions. In addition, we focus on how to apply the correlation to optimize the original candidate region features and improve video action detection accuracy. The experimental results show the promising improvement achieved by our method over the state-of-the-art solutions.     

基于空间光调制器衍射特性的匹配滤波器设计

利用空间光调制器衍射特性,在匹配滤波平面上对目标图像傅里叶谱的0级,±1级衍射谱同时加载3幅匹配滤波器,以拓宽覆盖待识别目标图像的畸变范围.根据设计方案,0级谱位置加载复合匹配滤波器,±1级衍射谱位置加载非复合匹配滤波器.针对特定目标进行了模拟和系统测试实验,证明了该方法有效可行.     

多源航迹相关准则探讨

雷达组网数据融合处理大致可分为信息提取、数据传输、信息分类、时空对准、航迹相关、数据综合等几个过程。探讨判断了来源于多部雷达的航迹之间相关的准则,针对两坐标搜索雷达航迹信息相关提出了相关流程和准则设计时需要考虑的几个方面内容。     

基于区域相关性的LSB匹配隐写分析

LSB匹配通过随机地修改像素值的LSB,打破了LSB替换的值对关系,使得安全性有了质的提高.针对灰度图像空间域LSB匹配算法,以8邻域内像素与中心像素的像素值差值定义区域内像素之间的相关性,分析嵌入信息对相关性的影响,提出了基于区域相关性的检测算法,与已有算法相比,该算法复杂度低,运算简单并且误检率低,当嵌入率高于0.3时检测结果具有较高的可靠性.     

基于小波多尺度相关的特征边缘提取方法

根据采用小波变换提取图像边缘的多尺度特性,对小波多尺度间的相关性进行了分析,提出多尺度小波变换相关算子,分析了相关算子的滤波特性,利用相关算子对图像的多尺度边缘信息进行了优化.既保留了图像中重要的细节边缘信息,又剔除大量的冗余边缘和虚假边缘,有效地提取出了图像的特征边缘.     

低信噪比下相关干涉仪测向处理方法

针对在低信噪比条件下相关干涉仪测向来波方位角估计误差较大甚至有时失效的问题,提出一种基于相位差矢量累积的相关干涉仪测向处理方法。与直接使用相位差的常规相关干涉仪测向不同,该方法首先利用多次测量的相位差和信号幅度构造复数矢量,然后对这些矢量进行累积运算,再与天线的阵列流形进行拟合。此方法通过提高相位差的测量精度和稳定度从而达到即使在低信噪比条件下也可以得到较高测向准确度,具有算法简单、性能稳健的特点。最后通过仿真分析,对文中提出的方法给予验证。     

Design and analysis of low-power body area networks based on biomedical signals

A body area network (BAN) as one branch of Sensor Networks, is an inter-disciplinary area which holds great promises for revolutionising the current health care systems. BAN combines the real-time updating of biomedical data with the continuous and dynamic health care monitoring closely. A number of intelligence biomedical sensors can be integrated into a wireless BAN system, and the system can be used for prevention, diagnosis and timely treatment of various medical conditions. In this article, we propose a data fusion technique for a BAN based on biomedical signals. This proposed solution is of much lower complexity than conventional techniques and hence can significantly reduce the power consumption in the BAN. The technology is carried out by removing redundant and unnecessary sample information and shifting a large portion of processing and control loads to the remote control centre in an asymmetric manner. This approach not only reduces the power consumption of biosensor nodes in a BAN, but also ensures the integrity of the biomedical information. In addition, we present a self-designed distributed time–space correlation compressive sensing model and propose an efficient algorithm based on biomedical signals. Simulation results show that the proposed algorithm can not only reconstruct the original signal with high accuracy and but also achieve significant reduction in power consumption.     

语音系统客观音质评价研究

 本文采用以谱距离为基础的LPC-CD、Bark-SD和Mel-CD三种客观失真测度,对若干种语音编码设备及它们的级联系统进行了客观音质评价实验研究.结果表明,这三种测度单独应用都具有较好主、客观相关性,它们的组合,给出更好的相关结果.因此,可以推断,这些测度作为语音系统的音质评价是有一定有效性和实用性的.     

基于压缩矩阵Apriori算法的高校学生成绩相关性分析研究

针对高校学生成绩数据库的特点,采用基于压缩矩阵的Apriori算法来分析学生各学科成绩之间的相关性。该算法可以使矩阵更小,并减少扫描数据库和压缩矩阵的次数。通过分析学生各学科成绩的关联,找出有意义的关联规则,可以为教师进行教学活动及教学管理人员制定教学计划、进行教学管理等提供参考。     

二进制粒属性相关判定与应用

采用二进制粒表示信息系统中的离散化属性值,即对属性进行二进制粒化,定义了基于二进制信息粒的匹配、合取及析取算子,描述了二进制粒属性相关的定义,提出了基于二进制粒的属性相关性判定矩阵的算法,算法通过判定信息系统中属性的相关性,为属性约简及关联规则的求解提供一种新方法.如果相关属性同为条件属性,则可进行约简,选择其一;如果有决策属性,则可生成强关联规则;该算法在气象数据的仿真实验中验证了其实用性.