关于聚类有效性函数熵公式HP（u,c）

对文[1]提出的聚类有效性函数HP(u,c）作了一定的理论分析,并就文[1]使用的数据及其他数据进行了计算机模拟。模拟结果显示：HP（u,c) 作为FCM算法的聚类有效性函数是不合适的。     

信息熵方程求解算法及其应用

给定一个离散且有限随机变量的信息熵,求其对应的概率分布需要解多元非线性方程,文中提出了一个将n元信息熵方程化为至多(n-1)个一元非线性方程求解的算法,证明了算法的正确性,给出了算法误差估计;运用熵方程求解算法设计了一种基于信息熵的文本数字水印方案.     

信息熵在粗糙集信息检索模型中的应用

在信息检索过程中,由于文档中存在大量的多义和近义现象,导致不确定性出现,这将影响检索的性能.为此我们采用信息熵和粗糙集理论来处理这类不确定性问题.首先计算训练文档集中的词之间的信息熵,对信息熵做模糊聚类来构造词之间的等价关系,然后借助于该等价关系提出并实现了一个以粗糙集上下近似为基础的信息检索模型,通过实验的测试,该模型能够提高信息检索的效率.     

Vague集信息熵测量及其应用

引入了区别于现有文献的Vague集信息熵和Vague集的关联熵的概念,给出了一种改进的测量方法,并讨论了它们之间的关系。进而,我们揭示了Vague集的熵和Fuzzy集的熵之间的关系,并分析了本文所定义熵的意义。最后,讨论了这种关联熵在模糊识别和医疗诊断上的应用。     

信息熵公理及其在决策分析中的应用

针对信息量是消息发生前的不确定性给出一个直观测量信息量公式.为了克服Shannon熵的局限性和分析信息度量本质,借鉴距离空间理论中度量公理定义的思路,通过非负性、对称性、次可加和极大性给出信息熵的公理化新定义.将Shannon熵、直观信息熵和β-熵等不同形式的信息度量统一在同一公理化结构下.应用直观信息熵公式仅采用四则运算进行决策树分析,避免了利用Shannon熵公式的对数运算.     

标准差和平均差的内在关系及应用研究

由于绝对值不是变量的解析函数,不具有方差的良好数学性质;但由于它对极少数极端事件不敏感,因而更具有抗干扰性。在实证研究波动时,更偏向于平均差。本文正是在此矛盾的基础上,通过考虑分布和不考虑分布的两种情况,从理论和应用的不同角度,深入地研究了两者间的内在关系及应用情况,从而为科学合理地利用离散度指标提供新思路。     

关于分圆域及其最大实子域的类数的上界

设ｍ是适合ｍ≠２（ｍｏｄ４）的正整数，ζｍ是ｍ次本原单位根，又设△ｋ，ｈｍ分别是分圆域Ｋ＝Ｑ（ζｍ）的判别式和类数，本文证明了：当ψ（ｍ）≥２２０时，ｈｍ〈４２３ｗｍＱｍ√△ｋ／（１９．４７）其中ψ（ｍ）是ｍ的Ｅｕｌｅｒ函数ｗｍ是Ｋ中单位根的人数，当ｍ是素数方幂时，Ｑｍ＝１否则Ｑｍ＝２由此可推知：当奇素数Ｐ≥２２３时，ｈｐ〈３６ｐ＾７．５（ｐ／２１．６）＾（ｐ－２）／２     

关于Tsallis熵与随机序的一些结果

本文讨论了具有相同Tsallis熵或相同Tsallis相对熵的两个连续随机变量随机等价的一些条件,以及随机序与Tsallis熵序的一些关系.     

关于Tsallis型非对称熵的若干研究

基于Tsallis熵和非对称熵,本文提出了Tsallis型非对称熵,该熵推广了Tsallis熵和非对称熵,证明了最大的Tsallis型非对称熵原理,并且从该原理中可以获得比Tsallis熵及非对称熵原理更多的分布,从而说明该原理的有用性.     

关于宽平稳随机过程的采样定理的一致收敛速度（Ⅱ）

继文献[1],求出了谱测度集中在[-π/△,π/△]上的具有连续参数的宽平稳随机过程x（t）的相关函数,谱密度函数和谱函数的估计及它们的一致收敛速度.这些估计及一些收敛速度都是基于离散采样（x（k△）,k=0,±1,±2,…）上的.     

On the efficiency and capacity of computers

We address the problems of estimating the computer efficiency and the computer capacity. We define the computer efficiency and capacity and suggest a method for their estimation, based on the analysis of processor instructions and kinds of accessible memory. Obtained results can be of some interest for practical applications.     

Bounds for Shannon and Zipf‐Mandelbrot entropies

Shannon and Zipf‐Mandelbrot entropies have many applications in many applied sciences, for example, in information theory, biology and economics, etc. In this paper, we consider two refinements of the well‐know Jensen inequality and obtain different bounds for Shannon and Zipf‐Mandelbrot entropies. First of all, we use some convex functions and manipulate the weights and domain of the functions and deduce results for Shannon entropy. We also discuss their particular cases. By using Zipf‐Mandelbrot laws for different parameters in Shannon entropies results, we obtain bounds for Zipf‐Mandelbrot entropy. The idea used in this paper for obtaining the results may stimulate further research in this area, particularly for Zipf‐Mandelbrot entropy.     

On the eigenvalue and Shannon's entropy of finite length random sequences

Pseudorandom binary sequences play a significant role in many fields, such as spread spectrum communications, stochastic computation, and cryptography. The complexity measures of sequences and their relationship still remain an interesting open problem. In this article, we study on the eigenvalue of random sequences, deduce its theoretical expectation and variance of random sequences with length N, and establish the relationship between eigenvalue and Shannon's entropy. The results show that these two measures are consistent. Furthermore, the eigenvalue of random n‐block sequences and its relation to Shannon's entropy are also been studied. © 2014 Wiley Periodicals, Inc. Complexity 21: 154–161, 2015     

基于极大熵准则的先验分布确定方法

借助于熵的概念，讨论了用极大熵的思想来确定先验分布的几种情形，给出了在各种情形下先验分布的形式和结论，从而提供了确定先验分布的一种有效的方法。     

Generalized Informational Entropy and Noncanonical Distribution in Equilibrium Statistical Mechanics

Based on the Jaynes principle of maximum for informational entropy, we find a generalized probability distribution and construct a generalized equilibrium statistical mechanics (ESM) for a wide class of objects to which the usual (canonical) ESM cannot be applied. We consistently consider the case of a continuous, not discrete, random variable characterizing the state of the object. For large values of the argument, the resulting distribution is characterized by a power-law, not exponential, asymptotic behavior, and the corresponding power asymptotic expression agrees with the empirical laws established for these objects. The -deformed Boltzmann–Gibbs–Shannon functional satisfying the requirements of the entropy axiomatics and leading to the canonical ESM for =0 is used as the original entropy functional. We also consider nonlinear transformations of this functional. We show that depending on how the averages of the dynamical characteristics of the object are defined, the different (Tsallis, Renyi, and Hardy–Littlewood–Pólya) versions of the generalized ESM can be used, and we give their comparative analysis. We find conditions under which the Gibbs–Helmholtz thermodynamic relations hold and the Legendre transformation can be applied to the generalized entropy and the Massieu–Planck function. We consider the Tsallis and Renyi ESM versions in detail for the case of a one-dimensional probabilistic object with a single dynamical characteristic whose role is played by a generalized positive energy with a monotonic power growth. We obtain constraints on the Renyi index under which the equilibrium distribution relates to a definite class of stable Gaussian or Levy–Khinchin distributions.     

连续型随机向量联合熵的离散方差分离估计

提出了一种"离散方差分离"法,用于连续型随机向量联合熵的估计.方法分为"方差分离"和"离散"两个步骤.前者通过分离"标准熵"与"标准差对数和"来避免维数灾害;后者通过各分量的"最佳分割数"来离散连续型随机向量,从而避开了联合密度估计.仿真实验表明:该方法以很低的计算复杂度,准确地逼近了理论值.     

On the Fine Structure of Stationary Measures in Systems Which Contract-on-Average

Suppose {f ₁,...,f _m } is a set of Lipschitz maps of ^d . We form the iterated function system (IFS) by independently choosing the maps so that the map f _i is chosen with probability p _i ( ^m _i=1 p _i =1). We assume that the IFS contracts on average. We give an upper bound for the upper Hausdorff dimension of the invariant measure induced on ^d and as a corollary show that the measure will be singular if the modulus of the entropy _i p _i log p _i is less than d times the modulus of the Lyapunov exponent of the system. Using a version of Shannon's Theorem for random walks on semigroups we improve this estimate and show that it is actually attainable for certain cases of affine mappings of .     

A Note on Some Inequalities for Finite Sums and an Application to Entropies of Probability Distributions

In this paper we prove some inequalities for finite sums with positive terms. As a consequence of these results we obtain an inequality for entropies of discrete probability distributions.