基于FCM的动态结合全局图像阈值分割

全局阈值分割对于小目标物效果不理想,动态阈值容易产生阴影等干扰,但综合考虑全局阈值和动态阈值可以达到比较理想的结果。模糊C均值算法用于灰度图像分割是一种非监督模糊聚类后再标定的过程,该文在不明显增加运算量的前提下,利用模糊C均值自动聚类的功能分别得到全局阈值和动态阈值,完成对阈值矩阵的构造和图像的分割。     

动静态属性数据相结合的客户分类方法研究

研究了一种客户动态、静态属性数据相结合的客户分类方法。提出了客户时间序列的加权处理方法,并应用客户时间序列的统计特征作为聚类特征向量,采用混合式遗传算法对客户聚类,使每一类客户具有相似的时序特征。在此基础上将聚类结果与客户的静态属性数据相结合,对客户进一步分类。实验结果表明,与传统的基于静态属性数据的客户分类方法相比,本文的方法提高了客户分类的准确性。     

Fuzzy K-means clustering models for triangular fuzzy time trajectories

We focus our attention on the classification of fuzzy time trajectories with triangular membership function, described by a given set of individuals. To this purpose, we adopt a fullyinformational approach, explicitly recognizing the informational nature shared by the ingredients of the classification procedure: the observed data (Empirical Information) and the classification model (Theoretical Information). In particular, by supposing that the informational paradigm has a fuzzy nature, we suggest three fuzzy clustering models allowing the classification of the triangular fuzzy time trajectories, based on the analysis of the cross sectional and/or longitudinal characteristics of their components (centers and spreads). Two applicative examples are illustrated.     

基于Web使用挖掘的个性化服务系统

个性化服务系统是一种由多种WEB挖掘技术构成的基于用户使用的站点个性化系统。该系统使用事务聚类、使用聚类和关联规则技术等数据挖掘技术分析用户访问模式,并结合用户当前访问情况提供实时化个性服务。实验结果说明,个性化服务系统具有较好的性能。     

A bayesian estimator for the dependence function of a bivariate extreme‐value distribution

Any continuous bivariate distribution can be expressed in terms of its margins and a unique copula. In the case of extreme‐value distributions, the copula is characterized by a dependence function while each margin depends on three parameters. The authors propose a Bayesian approach for the simultaneous estimation of the dependence function and the parameters defining the margins. They describe a nonparametric model for the dependence function and a reversible jump Markov chain Monte Carlo algorithm for the computation of the Bayesian estimator. They show through simulations that their estimator has a smaller mean integrated squared error than classical nonparametric estimators, especially in small samples. They illustrate their approach on a hydrological data set.     

关于参数化测度（Ⅱ）

引进一类拓扑向量空间，用以描述测度的偏导。使用的工具是所谓的参数化鞅。建立了若干参数化鞅的收敛定理。     

甘肃省县域经济发展水平空间差异分析

甘肃省包括86个县区,受地理位置、气候条件、资源状况、经济基础各种因素的影响,各县区之间的经济发展水平差异显著但又有很多相似之处。为了定量揭示甘肃省县域经济发展水平在空间上的分布现状、差异以及相似性,笔者拟运用多元统计方法对甘肃省县域经济发展水平的空间差异分析     

基于复杂网络的投资组合优化研究

本文基于复杂网络的局部聚类系数改进了传统的全局最小方差投资组合模型。首先通过股票对数收益率的相关系数矩阵构造股票关联网络,然后计算股票关联网络的局部聚类系数,最后通过全局最小方差模型确定最佳投资组合。将改进后的模型应用于A股市场,经过夏普比率、信息比率和欧米茄比率的对比分析得出改进后的投资组合模型在样本外的表现优于传统的全局最小方差投资组合模型。     

One-dimensional optimal bounded-shape partitions for Schur convex sum objective functions

Consider the problem of partitioning n nonnegative numbers into p parts, where part i can be assigned n_i numbers with n_i lying in a given range. The goal is to maximize a Schur convex function F whose ith argument is the sum of numbers assigned to part i. The shape of a partition is the vector consisting of the sizes of its parts, further, a shape (without referring to a particular partition) is a vector of nonnegative integers (n₁,..., n_p) which sum to n. A partition is called size-consecutive if there is a ranking of the parts which is consistent with their sizes, and all elements in a higher-ranked part exceed all elements in the lower-ranked part. We demonstrate that one can restrict attention to size-consecutive partitions with shapes that are nonmajorized, we study these shapes, bound their numbers and develop algorithms to enumerate them. Our study extends the analysis of a previous paper by Hwang and Rothblum which discussed the above problem assuming the existence of a majorizing shape. This research is partially supported by ROC National Science grant NSC 92-2115-M-009-014.     

信息不完全确定的区域产品模糊区间聚类方法

区域产品分类与选择是区域经济发展中最重要和最基础的工作.在产品选择与分类时,需要确定指标权系数和分类阈值等参数,这在实际应用中是比较困难的.针对这种情况,提出了一种信息不完全确定的区域产品模糊区间聚类方法.该方法构建了指标权系数信息不完全确定的最优模糊区间聚类模型,利用遗传算法和改进的FCM算法联合求解所得优化模型,得到指标权系数、最优聚类中心和最优划分,进而确定各产品所属类别.最后将该方法应用于某区域的产品分类和主导产品的确定中,实例计算说明该方法的可行性和有效性.