关于“信息熵”分析及其应用的研究

本文用Ｓｈａｎｎｏｎ熵函数的观点及最大离散熵原理，探讨了对随机试验项目的优选问题：即压缩某种“不确定性”试验项目的实施与探索．并通过应用实例的计算，获得了满意的聚类效果．这种方法对于信息资料的数据处理及压缩“不确定性“验项目具有普遍意义．     

Clustering Effect on the Statistical Estimation Accuracy of Distribution Density

The paper is devoted to statistical nonparametric estimation of multivariate distribution density. The influence of data pre-clustering on the estimation accuracy of multimodal density is analyzed by means of the Monte Carlo method. It is shown that the soft clustering is more advantageous than the hard one. While a moderate increase in the number of clusters also increases the calculation time, it considerably reduces the estimation error.     

The Huber theorem for non-compact conformally flat manifolds

It was proved in 1957 by Huber that any complete surface with integrable Gauss curvature is conformally equivalent to a compact surface with a finite number of points removed. Counterexamples show that the curvature assumption must necessarily be strengthened in order to get an analogous conclusion in higher dimensions. We show in this paper that any non compact Riemannian manifold with finite -norm of the Ricci curvature satisfies Huber-type conclusions if either it is a conformal domain with volume growth controlled from above in a compact Riemannian manifold or if it is conformally flat of dimension 4 and a natural Sobolev inequality together with a mild scalar curvature decay assumption hold. We also get partial results in other dimensions. Received: April 14, 2000; revised version: March 20, 2001     

On a new stochastic global optimization algorithm based on censored observations

In this paper a new algorithm is proposed for global optimization problems. The main idea is that of modifying a standard clustering approach by sequentially sampling the objective function while adaptively deciding an appropriate sample size. Theoretical as well as computational results are presented.     

煤与瓦斯突出预报的模糊聚类关联分析法

首先采用模糊聚类分析对煤与瓦斯突出的样本集合进行分类，建立不同突出程度的模糊模式。然后用关联分析确定待预报样本与模式的关联程度，以此预测预报样本的煤与瓦斯突出危险程度。这种预报方法相对于模糊聚类分类后，将模式与待预报样本组成新的样本集合，再进行聚类分析，以此分类结果进行预报法。不仅可靠程度高，而且能定量描述待报样本与模式的亲和程度。     

The k-means range algorithm for personalized data clustering in e-commerce

This paper describes the k-means range algorithm, a combination of the partitional k-means clustering algorithm with a well known spatial data structure, namely the range tree, which allows fast range searches. It offers a real-time solution for the development of distributed interactive decision aids in e-commerce since it allows the consumer to model his preferences along multiple dimensions, search for product information, and then produce the data clusters of the products retrieved to enhance his purchase decisions. This paper also discusses the implications and advantages of this approach in the development of on-line shopping environments and consumer decision aids in traditional and mobile e-commerce applications.     

关于Fuzzy聚类分析及其在脊髓灰质炎中的应用

本文用Fuzzy数学的观点探讨了Fuzzy聚类分析，并通过对脊髓灰质炎病毒抗体水平的Fuzzy聚类分析，获得了良好的聚类效果，为该项问题的研究提供了科学依据。     

Identifying potential binding modes and explaining partitioning behavior using flexible alignments and multidimensional scaling

A method is described for the rapid and automatic analysis of flexible molecular alignments using multidimensional scaling and a normalized scoring scheme. A projection scheme was devised to separate orientational and conformational effects. It is shown that the approach can be utilized for the identification of common binding orientations or to the study of differences in partioning behavior. It is suggested that the method can be employed as a novel approach exploring molecular similarity as a dynamic property, so that it includes aspects of motion (by way of mutual orientations), conformations and molecular properties.     

A Method for Clustering and Screening of Long-dimensional Chemical Data Based on Fingerprints and Similarity Measurements

A method for the treatment of long-dimensional chemical data arrays is presented in this work with the aim of maximising classification models. The method is based on the construction of fingerprints and the subsequent generation of a similarity matrix. The similarity calculation has been modified through a scaling process to take into account different significance shown by the variables. The method was applied to spectral measurements of wines and several aspects were studied, namely: threshold considered in the construction of fingerprints and patterns, weighting factor used for scaling, normalisation method, etc. The application of both Principal Components Analysis and Soft-Independent Modelling of Class Analogies to the similarity matrices gave better classifications of the information than those obtained using original data.     

Partial least-squares regression and fuzzy clustering — A joint approach

Chemical and physical analyses of malt, the main ingredient of beer, have been used to predict the concentration of certain volatile compounds in the finished beer.The prediction was done by means of the partial least squares regression (PLS2) in SIMCA. The total data set as well as individual malt clusters were submitted to PLS analysis. Best prediction was obtained by separating the total object matrix in classes according to similarity found by fuzzy pattern recognition (FCV). FCV was also used to separate the beer variables in classes and to select the subset of variables to be predicted.A joint approach of fuzzy pattern recognition to identify groups of samples and SIMCA-PLS2 to predict several dependent variables is suggested as a powerful tool in process-analytical chemistry.