Rough law energy and measurement of ¯F-decomposition rough law

By employing function one-direction S-rough sets and rough law generation method based on function S-rough sets, ¯ f-decomposition law and ¯ F-decomposition rough law are proposed, and the measurement of rough law variation in the process of rough law ¯ F-decomposition is researched. The concepts of law energy and attribute ¯ f-interference degree are presented, which make the variation of rough law become measurable. ¯ f-decomposition law energy characteristic theorem, ¯ fdecomposition law energy inequality theorem, ¯ F-decomposition rough law energy characteristic theorem, and ¯ f-decomposition law energy mean value theorem are presented.     

ON ALGEBRICO-DIFFERENTIAL EQUATIONS-SOLVING

The char-set method of polynomial equations-solving is naturally extended to the differential case which gives rise to an algorithmic method of solving arbitrary systems of algebrico-differential equations. As an illustration of the method, the Devil's Problem of Pommaret is solved in details.     

论生物学哲学与传统科学哲学间的张力

本文概述生物学哲学的研究传统,进而讨论生物学拥有对于在传统科学哲学里的问题,包括自然类,发现和确证,解释以及还原等的暗示是什么,其中的某些暗示提供了对于传统形而上学、语义学的认识论观点的新阐述,并且阐明了进行哲学研究的多种途径,这超出了大多数传统哲学的实践,并指出生物学哲学的自然主义方法论启示。     

基于PVS的UML类图和序列图的一致性检验

针对UML类图和序列图的一致性问题,在充分考虑了类继承关系、关联关系、类方法的可见性以及类方法的前、后置条件等因素对一致性影响的基础上,给出了判定类图和序列图一致性的必要条件和PVS元理论,提出了一种基于定理证明器PVS的一致性检验方法.在检验UML模型一致性时,把一致性检验问题转化为逻辑定理证明问题.实践表明,该方法对于提高UML模型的可信度,减少系统实现阶段的错误起到了一定作用.     

基于频谱聚类分割的粒子群最优模态分解仿真研究

在两种有力措施的基础上提出了粒子群最优模态参数识别算法.一是提出了一种性能稳定的模态参数初始值估计算法,引入模态聚类的思路来估计出各个模态参数的上下限范围.该算法把幅谱曲线看成是局部波峰的集合,按聚类分割思路来构造聚类距离函数,使用k-means算法把振动信号频谱自动聚类成多个单模态类,然后运用单模态分解算法估计出每个模态类的模态参数的上下限范围,给出粒子属性值的上下界,极大地减少粒子群算法的搜索空间,减少最优搜索时间提高搜索结果的稳定性.二是采用了混合变异粒子群算法来提高最优化搜索的效率,有效避免陷入局部最优,提高模态参数的准确性.从仿真信号的大量实验研究结果看,与经典的正交多项式拟合算法相比,该算法的噪声抵抗能力更强、更稳定.     

New incremental clustering framework based on induction as inverted deduction

A new incremental clustering framework is presented, the basis of which is the induction as inverted deduction. Induction is inherently risky because it is not truth-preserving. If the clustering is considered as an induction process, the key to build a valid clustering is to minimize the risk of clustering. From the viewpoint of modal logic, the clustering can be described as Kripke frames and Kripke models which are reflexive and symmetric. Based on the theory of modal logic, its properties can be described by system B in syntax. Thus, the risk of clustering can be calculated by the deduction relation of system B and proximity induction theorem described. Since the new proposed framework imposes no additional restrictive conditions of clustering algorithm, it is therefore a universal framework. An incremental clustering algorithm can be easily constructed by this framework from any given nonincremental clustering algorithm. The experiments show that the lower the a priori risk is, the more effective this framework is. It can be demonstrated that this framework is generally valid.     

加权快速聚类距离解模糊算法

距离模糊是雷达系统中重频工作模式下必须考虑的问题,而多重频技术是解距离模糊常见的信号波形设计方案。一维聚类算法可根据雷达不同重频的测量视在距离稳健地求解目标不模糊距离,但一维聚类算法在排序效率和根据测距信噪比估计目标不模糊距离性能两方面存在不足。加权快速聚类距离解模糊算法首先提出快速聚类算法提高解模糊时的排序效率,继而采用加权方式提高目标不模糊距离的估计性能。快速聚类算法的仿真试验结果表明快速聚类算法解距离模糊是一种实用的快速解距离模糊算法。     

外P-集合与F-信息伪装

P-集合（packet sets）是一个集合对,它是由内P-集合（internal packet sets）与外P-集合（outer packet sets ）共同构成。P-集合具有动态性,利用外P-集合,给出F-信息伪装与F-信息伪装生成概念;提出F-信息伪装度量定理与信息伪装被恢复〖CD*2〗还原定理;给出辨识准则与方法。利用这些结果,给出F-信息伪装被恢复还原的应用。P-集合是研究动态信息系统的一个新理论和新方法。     

基于分解-聚类-集成学习的汇率预测方法

本文集成了经验模态分解(EEMD)、最小二乘支持向量回归(LSSVR)和K均值聚类方法,提出了一个新的外汇汇率预测方法,称为基于EEMD-LSSVR-K的分解-聚类-集成学习的外汇汇率预测方法.该方法利用聚类策略将分解-集成学习中固定权值集成学习扩展到基于局部数据特征加权的非线性集成加权学习,从而克服了分解-集成方法中集成学习阶段的不足.本文将该方法用于四种主要外汇汇率的预测,实证结果表明:在提前1天、提前3天和提前6天的预测中,本文所提出的EEMD-LSSVR-K方法的水平预测性能和方向预测性能显著地优于基准模型;同时也证实了聚类策略能够有效提高分解-集成模型的预测效果.     

改进投票策略的多类SVM及在故障诊断中应用

针对一对一(OVO)分解法,提出了一种改进的投票(MWV)策略,解决了传统策略中的不可分区域问题。首先,由训练ω_i类和ω_j(j≠i,j=1,…,n)类而得到的SVM决策函数;再对ω_i类定义了一个取值在0~1之间的调节函数,并使改进的得票值等于传统得票值加上调节函数。最后,根据改进的得票值进行分类决策。对于可分区域的样本,改进MWV策略的分类结果与传统策略完全相同;对于不可分区域的数据,由调节函数的值决定。将所提法应用于齿轮传动箱故障诊断实例并与传统得票策略诊断进行了对比,实验结果验证了所提方法的上述优越性。     

Function S-Rough sets and its applications

1 .INTRODUCTIONBased on S-rough sets(singular rough sets)[3 ~14],Refs .[1 ,2] presented function S-rough sets (func-tion singular rough sets) and its two forms :func-tion one direction S-rough sets (function one direc-tion singular rough sets) and function two direc-tion S-rough sets (function two direction singularrough sets) . Function S-rough sets is obtained toresearch mining discovery . Let’s see a practicalsystem: The output state of a system can be de-scribed by the function set…     

单向S-粗集对偶与数据筛选-过滤的扩展应用

S-粗集(singular rough sets)是对Z.Pawlak粗集的改进,单向S-粗集对偶(dual of one direction sin-gular rough sets)是S-粗集的基本形式之一。利用单向S-粗集对偶,给出数据属性,数据筛选-过滤概念,数据筛选-过滤序定理,合成数据筛选-过滤定理,及数据筛选-过滤准则。利用这些结果,给出应用。单向S-粗集对偶是动态数据筛选-过滤研究的一个新工具。     

规律与它的f-属性控制

函数单向S-粗集对偶(dual of function one direction singular rough set),具有单向动态特性和规律特性;它是函数S-粗集(function singular rough set)的基本形式之一。函数S-粗集是在改进S-粗集的基础上提出的。利用函数单向S-粗集对偶的动态特性和规律特性,给出f·-规律,f·-规律的属性特征,属性距离,f·-冗余规律概念。利用这些概念,提出规律与它的f·-属性控制,并给出f·-属性控制定理,f·-属性控制判定定理,f·-属性控制识别准则与应用。     

Function S-rough sets and mining-discovery of rough law in systems

1. INTRODUCTION S-rough sets (singular rough sets) was presented in Ref. [3] in 2002, and defined on α -element equival- ence class [x] with dynamic characteristic. S-rough sets has more advantages than Z.Pawlak’s rough sets[24]. S-rough sets has been applied in dynamic object recognition[21], mechanical engineering[22] and information science[8-13]. In 2005, function S-rough sets was put forward in Refs. [1,2], and is defined on α -function equivalence class [u] with dynamic charact…     

F−rough law and the discovery of rough law

By using function one direction S-rough sets (function one direction singular rough sets), this article presents the concepts of F-law, F-rough law, and the relation metric of rough law; by using these concepts, this article puts forward the theorem of F-law relation metric, two orders theorem of F-rough law relation metric, the attribute theorem of F-rough law band, the extremum theorem of F-rough law relation metric, the discovery principle of F-rough law and the application of F-rough law.     

S-粗集与F-隐藏知识发现

S-粗集具有三类形式:单向S-粗集,双向S-粗集,单向S-粗集对偶。S-粗集具有动态特性,遗传特性,记忆特性。利用单向S-粗集对偶与它的隐藏特性,本文给出f-隐藏知识,F-隐藏知识,隐藏度,隐藏依赖的概念,提出隐藏知识的隐藏定理,隐藏知识的隐藏依赖定理,给出F-隐藏与F-隐藏依赖在系统状态识别中的应用。     

单向S-粗决策规律与决策规律挖掘

利用单向S-粗集,给出单向S-粗决策规律生成方法;给出上决策规律,下决策规律,单向S-粗决策规律核,单向S-粗决策规律带,单向S-粗决策规律壳的概念;利用这些概念,提出下决策规律传递定理,上决策规律传递定理,F-分离的属性定理,粗决策规律挖掘定理,与粗决策规律挖掘准则。     

Rough law generation and its separation-recognition

By using function S-rough sets (function singular rough sets), this paper gives rough law generation and the theorem of rough law generation. Based on these results above, the paper proposes rough law separation, the theorem of rough law separation, the compound generation theorem of rough law bands, and the principle of rough law bands. In the end, an application of rough law separation in recognizing the risk law of profit is presented.     

S-粗集与新材料发现-识别

S-粗集(singular rough sets)具有两类形式:单向S-粗集(one direction S-rough sets)和双向S-粗集(two direction S-rough sets)。S-粗集具有遗传特征、记忆特征。把S-粗集与材料科学进行学科渗透,互补共享,给出新金属材料的发现-识别的讨论,利用属性生成模型给出新金属材料的属性值分析,给出的结果与实际相符。S-粗集是粗集研究的一个新方向,是新材料发现-识别的一个新的数学工具。     

S-rough sets and the discovery of F-hiding knowledge

Singular rough sets (S-rough sets) have three classes of forms: one-directional S-rough sets, dual of one-directional S-rough sets, and two-directional S-rough sets. Dynamic, hereditary, mnemonic, and hiding properties are the basic characteristics of S-rough sets. By using the S-rough sets, the concepts of f-hiding knowledge, F-hiding knowledge, hiding degree, and hiding dependence degree are given. Then, both the hiding theorem and the hiding dependence theorem of hiding knowledge are proposed. Finally, an application of hiding knowledge is discussed.