H原子对固态合金化颗粒表面成份分布影响的研究

本文研究了氢化物ＴｉＨ２对Ｆｅ＋Ｔｉ机械合金化非晶储氢合金的表面成份偏聚和性能的影响。应用Ｘ射线光电子谱（ＸＰＳ）和俄歇电子谱（ＡＥＳ）对样品表面由表及里逐层测量成份分布。同时应用热重分析方法研究了样品氧化性能和产物。研究结果表明，在机械合金化过程中，Ｈ原子不仅能使ＦｅＴｉ非晶化，同时促使表面产生明显的Ｆｅ原子偏聚。表面Ｆｅ／Ｔｉ原子比接近７。非晶ＦｅＴｉ（Ｈ）相的氧化分两个阶段进行，Ｔｉ原子首先氧化（８３３Ｋ），随后Ｆｅ原子氧化（８９０Ｋ）。     

划分IC缺陷团的聚类算法

大量实验表明了在集成电路（IC）制造过程中形成中的缺陷团之间具有很强的相关性，然而现用于划分缺陷团的方法均忽略了缺陷之间的相关性。因此，得到了缺陷分布规律不能有效地反映缺陷在圆片上的分布，为了提高IC成品率和可靠性仿真和设计的精度，提高缺陷分布模型的准确性，该文提出了划分缺陷团的聚类算法，该方法依据缺陷形成的动力学基础，充分考虑了缺陷团之间的相关性，实验证明该算法用于划分IC的缺陷团是十分有效的。     

Cluster analysis in industrial market segmentation through artificial neural network

Market segmentation has commonly applied cluster analysis. This study intends to make the comparison of conventional two-stage method with proposed two-stage method through the simulated data. The proposed two-stage method is the combination of self-organizing feature maps and K-means method. The simulation results show that the proposed scheme is better than the conventional two-stage method based on the rate of misclassification.     

Machine assignment and part-families formation using group technology

The conventional way of solving the group technology (GT) problem is to start from an assignment of parts to machines and try to find a partitioning of machine cells and part families. The similarity between parts is measured based on commonality of the machines assigned to them. However, parts are assigned to machines based on their operation requirements and the operation capabilities of machines. Similarity between parts should be based on their required operations. In this paper, the authors attempt to solve or facilitate solving the GT problem at the assignment level. An algorithm for assigning parts to machines is provided which utilizes the types of operations required by parts and applies GT principles in producing the assignment. This leads to better partitioning of machine cells and part-families. Furthermore, operation sequences required by parts in determining the similarity between parts have been considered. An algorithm to form part-families based on the operation sequence similarity coefficient has been developed. The resulting families are then used by the assignment algorithm to produce machine assignments to part-families. The use of the algorithm is demonstrated by examples.     

Software for dynamic reconfigurable order picking systems

This paper describes an approach, conceptual framework, and software architecture for dynamic reconfiguration of the order picking system. The research and development project was sponsored by the Material Handling Research Center (MHRC), a National Science Foundation sponsored Cooperative Industry/University Research Center. The storage configuration is assumed to be an in-the-aisle order picking system in which stockkeeping units (SKUs) can occupy variable capacity storage locations and stock-splitting is allowed among zones (clusters). The product mix may include multiple product families with different life cycles, correlated demand within families and commonality of demand across families.     

区域内旱情监测站点的布设研究

本文对北京通州辖区范围内组织的大规模的土壤墒情调查数据进行总结。基于地理统计、空间分析以及统计学知识，分析了北京通州旱情分布形势。运用数值分类学的知识对测量数据进行了聚类分析，同时用DB(DaviesBouldin)指标准则研究了区域内合理布设旱情监测站的数量及位置，为建立旱情监测站点提供了理论依据。     

On denoising and compression of DNA microarray images

We address the problems of noise and huge data sizes in microarray images. First, we propose a mixture model for describing the statistical and structural properties of microarray images. Then, based on the microarray image model, we present methods for denoising and for compressing microarray images. The denoising method is based on a variant of the translation-invariant wavelet transform. The compression method introduces the notion of approximate contexts (rather than traditional exact contexts) in modeling the symbol probabilities in a microarray image. This inexact context modeling approach is important in dealing with the noisy nature of microarray images. Using the proposed denoising and compression methods, we describe a near-lossless compression scheme suitable for microarray images. Results on both denoising and compression are included, which show the performance of the proposed methods. Further experiments using the results of the proposed near-lossless compression scheme in gene clustering using cell-cycle microarray data for S. cerevisiae showed a general improvement in the clustering performance, when compared with using the original data. This provides an indirect validation of the effectiveness of the proposed denoising method.     

Non-redundant data clustering

Data clustering is a popular approach for automatically finding classes, concepts, or groups of patterns. In practice, this discovery process should avoid redundancies with existing knowledge about class structures or groupings, and reveal novel, previously unknown aspects of the data. In order to deal with this problem, we present an extension of the information bottleneck framework, called coordinated conditional information bottleneck, which takes negative relevance information into account by maximizing a conditional mutual information score subject to constraints. Algorithmically, one can apply an alternating optimization scheme that can be used in conjunction with different types of numeric and non-numeric attributes. We discuss extensions of the technique to the tasks of semi-supervised classification and enumeration of successive non-redundant clusterings. We present experimental results for applications in text mining and computer vision.     

Learning Recursive Bayesian Multinets for Data Clustering by Means of Constructive Induction

This paper introduces and evaluates a new class of knowledge model, the recursive Bayesian multinet (RBMN), which encodes the joint probability distribution of a given database. RBMNs extend Bayesian networks (BNs) as well as partitional clustering systems. Briefly, a RBMN is a decision tree with component BNs at the leaves. A RBMN is learnt using a greedy, heuristic approach akin to that used by many supervised decision tree learners, but where BNs are learnt at leaves using constructive induction. A key idea is to treat expected data as real data. This allows us to complete the database and to take advantage of a closed form for the marginal likelihood of the expected complete data that factorizes into separate marginal likelihoods for each family (a node and its parents). Our approach is evaluated on synthetic and real-world databases.