改进自适应遗传算法在关联规则中的研究

针对传统的关联规则在大数据环境下数据挖掘过程中效率低的问题,文中融入了遗传算法,并且对传统的遗传算法进行优化以便更好地适用于关联规则的挖掘。针对传统的自适应遗传算法,对其交叉算子和变异算子进行自适应优化,在此基础上加入兴趣度用来剔除一些没有实际效果的规则。实验结果表明在关联规则提取方面,结合云计算技术和所改进的算法可以优化其挖掘效率,并克服适应性和可用性等方面的缺陷,不仅加快了算法的收敛速度,也提高了解的质量。     

单入单出模糊控制器控制函数与插值式的偏差分析

 本文进一步分析了单入单出模糊控制器的插值机理,借助符号计算推导了模糊蕴涵算子分别选择取小算子和乘积算子时,单入单出模糊控制器控制函数的解析式;得到了控制函数与插值函数间的偏差表达式.最后本文给出了偏差的一个上界估计以及模糊控制器实际使用时确定规则数的公式.     

数据同化框架下红外与可见光图像的可控融合

针对现有图像融合方法的规则不易根据融合图像的后续使用要求进行自适应调整,并且各种方法的优点不易综合的问题,提出一种数据同化框架下的基于粒子群优化(PSD)和差分进化(DE)算法的融合方法。该方法把比率低通金字塔变换作为数据同化系统的模型算子,把非下采样的Contourlet变换(NSCT,nonsubsampled contourlet transform)作为观测算子,根据后续处理对图像各个属性指标值的依赖程度构造目标函数,再利用PSD和DE的混合算法来优化目标函数从而获取合适的影像。两组实验从视觉效果和定量指标两方面验证了方法的有效性。     

SKYLINE查询解析

现有的研究工作只考虑如何对单个底层关系表进行skyline计算,即它们假定用户所提交的skyline查询不涉及任何传统的关系操作,并且所有skyline维度均落入同一个关系表中.显然,在实际应用中,由于这种假设的不成立,使得在多数情况下用户查询的效率极其低下.基于此,将skyline计算作为一个特殊的关系操作符,研究它与传统关系操作符间执行顺序变换的等价规则.从而,利用这些等价变换规则,通过改变skyline操作符与传统关系操作符之间的执行顺序来有效提高查询的效率.同时,给出充分的理论证明来论证所给等价变换规则的正确性,并通过实验验证其有效性.     

Analysis of Distributed and Adaptive Genetic Algorithm for Mining Interesting Classification Rules

Distributed genetic algorithm can be combined with the adaptive genetic algorithm for mining the interesting and comprehensible classification rules.The paper gives the method to encode for the rules,the fitness function,the selecting,crossover,mutation and migration operator for the DAGA at the same time are designed.     

Technique for constructing grey-scale morphological operators usingfuzzy expert system

A new approach to grey-scale fuzzy mathematical morphology is presented. In this approach, any grey-scale window operator can be constructed using a fuzzy expert system. Many useful operators can be built using a small number of fuzzy rules     

命题逻辑系统SMTL中公式的积分真度理论

首先给出了强左连续t-模和SMTL命题逻辑系统的定义,证明了左连续的t-模为强左连续t-模当且仅当与之伴随的正则蕴涵算子为强正则蕴涵算子;其次,在基于强正则蕴涵算子的模糊命题逻辑系统中定义了公式的积分真度,给出了积分真度推理规则;最后,基于公式的积分真度在SMTL命题逻辑系统的全体公式集上引入了一种伪距离,提出了三种近似推理机制,从而使得在SMTL命题逻辑系统的统一框架下展开近似推理成为可能.     

基于矩阵二进制编码遗传算法的频繁项集挖掘

提出一种基于矩阵二进制编码的改进遗传算法MGA (Matrix Genetic Algorithm),应用于挖掘关联规则中的频繁项集。通过对初始种群的编码以及降维保证了合理的初始适应度,并对遗传算法中交叉算子和变异算子生成新个体与筛选的过程进行优化,使算法有优良的全局和局部搜索能力。实验结果显示,MGA算法的整体挖掘效率与质量良好。     

Fuzzy reasoning model based on universal combination operation model

The universal combination operation model is a comprehensive decision model of continuous-valued logic.It overcomes limitations of the scope of operations of reasoning operators in the current comprehensive decision-making system.This article discusses relationship of mutual information and general correlation coefficient and gives the corresponding rules of them,the optimal matching operator is selected to complete fuzzy decision according to mutual information between candidate attributes.The relationship of mutual information between attributes and generalized correlative coefficient provides the principle to select the matching operator.According to the results of experiment,it is more reasonable to enhance classified precision effectively.There is a certain application value of the article’s method.     

The radiation operator

The concept of the radiation operator is introduced to assist in the analysis of various problems involving sources and their radiation fields. It gives the field outside the source region as operating on the field of a point source. Because there is a simple connection between the radiation vector describing the far-field and the radiation operator, it can be used to define fields anywhere outside the source region from their values in the far-field zone. Another important properly of the radiation operator is its ability to express sources of fields given their radiation pattern and polarization in the far zone. The source of such a field can be written in the form of radiation operator operating on a current element, the delta function source. To interpret this in terms of computable functions, existing tables of operational rules for different classes of operators can be applied. Examples of radiation operators corresponding to different sources are given together with examples of sources corresponding to given radiation field patterns. Finally it is shown that the radiation operator allows a considerable simplification to the derivation of the multipole expansion theory when compared to the classical recursion-formula derivation through spherical harmonic eigenfunctions     

一种通信网络故障主动管理方法

为了弥补通信网络管理中仅接收网络事件进行故障管理的被动管理的不足,基于主动管理的思想,提出了健康检查法,并对此方法中关键的规则梳理、规则优化等进行了分析和总结,并给出了其实现方法,同时在某网络运营商的网络中进行了试验验证,获得了实际现网数据结果,得到了使用该方法可将运维人员从手工检查的重复劳动中释放出来、有效降低网络设备故障率的结论。     

Beating the duopoly with competition [telecommunications in UK]

The duopoly in UK telecommunications has ended now and competition in the industry is allowed within certain rules. The author examines the outcome of the Government's review and compares it to the IEE's response. The IEE proposes that the issuing of licences should take account of the ability to sustain viability as the sudden demise of an operator could leave customers with substantial problems     

基于欧式距离约束自适应遗传算法的参差码搜索

提出了一种基于欧式距离约束的自适应遗传算法（Euclidean distance-Adaptive Genetic Algorithm,EAGA）,该算法将欧式距离引入自适应交叉概率,使交叉概率随适应度和个体之间的相似度自适应变化,更好地增强种群的多样性,保存优良个体;为了防止EAGA在优化过程中出现退化现象,通过融合按照一定规则产生的新个体对算法进行了改进。采用EAGA选择最优参差比,使滤波器的零点尽可能的浅,在有效抑制杂波的同时避免目标丢失。同标准遗传算法相比,EAGA表现出了较好的搜索性能。     

Adaptive fuzzy process control of integrated circuit wire bonding

One step in the assembly of integrated circuits is wire bonding, requiring expert knowledge to optimize critical process characteristics. This paper describes a fuzzy logic controller which sets parameters for the wire bonding process for gold ball wire bonds, specifically controlling bonded ball diameter and shear strength density. While the focus is on control of ball bonds, the method is general and may be applied to other bonding methods with minor modifications. The design of the fuzzy engine and the development of rules based on both operator experience and textual sources are discussed. An adaptation technique is developed to tune the controller's membership functions based on the measured results of the process. The controller has been tested on actual devices in an industrial setting; results show improved process control in terms of consistency and reduced operator adjustment     

Increasing profitability and improving semiconductor manufacturingthroughput using expert systems

This paper describes a new procedure for using a machine-learning classification technique coupled with an expert system to increase profitability and improve throughput in a semiconductor manufacturing environment. The authors show how to use this procedure to identify relationships between work-in-process data (information obtained during semiconductor fabrication) and potential integrated circuit yield. The relationships, in the form of IF-THEN rules, are extracted from databases of previously fabricated integrated circuits and final yield. It is further shown that these rules, when incorporated into expert systems, can advise the human operator as to which batches of circuits are likely to produce submarginal yield if processed to completion, thereby providing a basis for developing or enhancing a quality control strategy. These rules also identify the parameters and values which have historically provided the highest and lowest final wafer yields. A cost analysis is given to illustrate the cost-effectiveness of this procedure. An introduction to semiconductor manufacturing and a glossary are provided     

A study of drilling performances with minimum quantity of lubricant using fuzzy logic rules

Nowadays the increasing interest to perform machining operations is in dry/near-dry environments. The reason includes health and safety of operator, cost, ease of chip recyclability, etc. However one important process, which is difficult to perform in dry, is drilling. Without coolant, drilling leads to excessive thermal distortion and poor tool life. In order to tackle these conflicting requirements, the essentiality of study on machining performances with minimum quantity lubricant (MQL) becomes important.Fuzzy logic rules, which are derived based on fuzzy set theory, are used to develop fuzzy rule based model (FRBM). The performance of FRBM depends on two different aspects: structures of fuzzy rules and the associated fuzzy sets (membership function distributions, MFDs). The aim of this study is to investigate the performances of FRBMs based on Mamdani and TSK-types of fuzzy logic rules with different shapes of MFDs for prediction and performance analysis of machining with MQL in drilling of aluminum alloy. A comparison of the model predictions with experimental results and those published in the literature shows that FRBM with TSK-type fuzzy rules describes excellent trade-off with experimental measurements.     

基于粗糙集的红外弱小目标检测方法

提出了利用粗糙集理论处理不完整信息的能力对红外弱小目标的运动轨迹进行辨识的检测方法。首先利用形态学滤波算法,分离出每帧图像中候选目标并提取其特征属性。把提取到的特征属性作为条件属性,把每帧图像中候选目标作为个体,构成知识决策系统。通过对决策表进行约简,得到决策系统的最小决策算法,并利用序列图像中目标运动的连续性和轨迹一致性来实现小目标的识别。实验表明,该方法能够有效解决目标跟踪过程中目标短暂丢失以及重现的问题,实现弱小目标的稳健检测和跟踪。     

通信运营商施工单位标准化管理研究

针对通信工程建设中施工单位管理存在的问题,本文将PMBOK项目管理理念与运营商现有的规章制度相结合,分析了运营商对施工单位管理的关键点,制定出一系列施工单位标准化管理动作,以统一不同施工单位的管理行为,提高通信工程建设水平。     

A fuzzy classifier with ellipsoidal regions for diagnosis problems

In our previous work, we developed a fuzzy classifier with ellipsoidal regions that has a training capability. In this paper, we extend the fuzzy classifier to diagnosis problems, in which the training data belonging to abnormal classes are difficult to obtain while the training data belonging to normal classes are easily obtained. Assuming that there are no data belonging to abnormal classes, we first train the fuzzy classifier with only the data belonging to normal classes. We then introduce the threshold of the minimum-weighted distance from the centers of the clusters for the data belonging to normal classes. If the unknown data is within the threshold, we classify the data into normal classes and, if not, abnormal classes. The operator checks whether the diagnosis is correct. If the incoming data is classified into the same normal class both by the classifier and the operator, nothing is done. But if the input data is classified into the different normal classes by the classifier and the operator, or if the incoming data is classified into an abnormal class, but the operator classified it into a normal class, the slopes of the membership functions of the fuzzy rules are tuned. If the operator classifies the data into an abnormal class, the classifier is retrained adding the newly obtained data irrespective of the classifier's classification result. The online training is continued until a sufficient number of the data belonging to abnormal classes are obtained. Then the threshold is optimized using the data belonging to both normal and abnormal classes. We evaluate our method using the Fisher iris data, blood cell data, and thyroid data, assuming some of the classes are abnormal