分类的

In the first approach the metadata extraction is viewed as a classification problem. Three statistic classification models, including Support Vector Machines, Perceptron with Uneven Margins, and Maximum Entropy are studied. The approach suffers from the problems of ignoring the variance in lengths of documents and ignoring the dependency of text units within documents. To solve the problems, our second approach takes the metadata extraction as a sequence data labeling problem and makes full use of the sequential information.

论文首先研究了基于分类的元数据抽取方法，考察了支持向量机、非对称间隔感知机和最大熵模型在ADME中的应用；然后针对分类方法中存在的忽略文档之间的差异、不考虑文档上下文等问题，提出了基于序列数据标注模型的抽取方法，并结合最大熵马尔可夫模型、投票感知机模型和条件随机场模型进行了研究；最后本文根据前两种方法的设计思想和优缺点，研究提出一种新的结构预测模型—— CDASP（Coordinate Descent Algorithm for Structure Prediction），并将其应用于元数据的自动抽取中。

This study is to give a solution for the estimation of flash point for mixtures, the necessity for classification of flammable liquids. The traditional models for predicting flash point of mixtures usually by the activity coefficient approach. However, the parameters of activity coefficient were regressed from phase equilibrium data in the literatures. If there is no such parameter in literatures for the desired mixture, the model cannot predict the flash point oft hat mixture. Thus, this research aims at the improvement of deficiency of the flash point prediction models in the literatures by useing UNIFAC (Universal Quasi-chemical Functional Group Activity Coefficient) equation, Dortmund-UNIFAC equation and Lyngby-UNIFAC equation. In this study, we were aim at the prediction model for miscible mixtures.From the result, it is suggested to use different type of UNIFAC equation to estimast activivity coefficient in the predition of flash point for different mixture type.

易燃液体分类的主要依据参数为闪火点值，因此本研究目的为解决易燃液体中混合物分类时所面临到的闪火点值缺乏的问题，而在过去文献中所提出的闪火点预测模式会使用到活性系数，但这些模式使用到的活性系数均需要相平衡的数据以回归其所需参数，若所欲估算闪火点的混合溶液文献上未有相关参数，则无法有效利用闪火点预测模式估算其闪火点，因此针对文献中现有闪火点预测模式的缺点，利用描述液体混合物活性系数UNIFAC (Universal Quasi-chemical Functional Group Activity Coefficient)方程式、Dortmund-UNIFAC方程式和 Lyngby-UNIFAC方程式，建立发展一闪火点预测模式，而本研究以互溶溶液为对象。

Experiments show that rhomb partitioning is more powerful in suppressing block artifacts than square partitioning, results in much higher reconstructed image quality, but takes longer encoding time; 2 For solving exhausting encoding problem of fractal image coding, we present a fast rhomb-partitioning fractal image coding based on neighbor-searching; we present a fast rhomb-partitioning fractal image coding based on block classification; we present a fast square-partitioning fractal image coding based on neighbor-searching, block classification and wavelet transform.

2针对分形图象压缩编码过程非常耗时的致命弱点，提出了基于邻域搜索的菱形块分割的快速分形图象压缩编码方法，提出了基于块分类的菱形块分割的快速分形图象压缩编码方法，提出了基于邻域搜索·块分类和小波变换的方块分割的快速分形图象压缩编码方法，实验结果表明，我们提出的快速分形图象压缩编码与传统的分形图象压缩编码相比，编码速度提高了近40倍。

Content characteristics of five kinds of detrimental elements,As,Cr,Cd,Pb,S,in iron ores imported from 23 countries,totally 490 batches,within recent two years were investigated by statistical techniques.Based on data distribution, interquantile robust statistics,and a brand-new data description technique,kernel density estimation,were used to acquire the standing values of detrimental elements.Multivariate statistical methods,including Principle Component Analysis and Factor Analysis,were used to ...

对我国进口铁矿石中的环境有害元素As、Cr、Cd、Pb、S的含量进行了总体统计分析，选用了四分位稳健统计分析及内核密度估计分别对铁矿中的有害元素As、Cr、Cd、Pb、S的含量的代表值进行估计，探讨了两种方法在代表值估计方面的特点，并使用主成分分析、因子分析等多元统计分析方法对有害元素的含量特征进行整体分类描述，确定了含量特征整体分类的主要有害元素，全面了解进口铁矿中涉及安全、环保等有害元素的质量状况。

As a result, some valuable taxonomicmethods, criteria and our viewpoints were concluded. The morphology ofthe sporus is the main identical base at species level, which include thecharacters of the shape, size, colour, septate and ornament of the wall etc,and some sporus with distinct variable are also useful at genus level.

孢子的特征(包括形状、大小、颜色、隔膜数和表面纹饰等)是种级分类的主要依据，一些形态差异较大的孢子可作为属的划分标准；其次，分生孢子梗的形态特征、产孢细胞的延伸方式及产孢方式也是主要的鉴别依据，对属级水平上该真菌的鉴别有重要意义；另外，培养特征及产孢表型可作为分类鉴定的必要、有益补充。

Secondly, in the research of fuzzy clustering methods, we gave out several useful formulae and an improved fuzzy transitive closure method according to the degrees of influence of various pyrolysis parameters. We also brought up a modified FCM algorithm according to the different functions of the samples.

其次，在对基于模糊聚类的识别方法研究中，根据各种热解参数在油气层预测中的影响程度，给出了构造模糊相似矩阵的几个有用的公式，以及改进的模糊聚类传递闭包法；根据每个样本点对分类的不同作用，提出了一种改进的模糊C均值算法；试验结果表明，这两种方法都可以有效地提高分类精度。

In this article, the study of Electroacoustic music was ...

第二部分从历史和技术的角度对电子音乐做了分类的研究，从早期的电子拼接音乐到电子音合成音乐再到当今的电子计算机音乐，这种分类的方式主要是从电子音乐的创作方式来考虑的。

Most divergences between classifications of the same substance should be removed over time either through co-operation between notifiers and registrants or by EU harmonized classifications for substances that are category 1, 2, and 3 CMRs, respiratory sensitizers, and other substances if justified on a case-by-case basis.

大多数对同种物质的分类的分歧须通过通报者或注册者之间的合作或由欧盟的统一分类即分I、II、III类CMRs物质，呼吸道感光剂及其他经个案一一证实的物质。

In this paper, the author discusses the origin and the development of the aseismic design medel code,approaches the role and the characteristics of the model code, points out the necessity and the ways of drawing up and developing the model code in China, and puts forward a set of relatively complete classification of engineering structure operation function based on the performance design, and proposes the principles, framework and the methods of aseismic design classification.

文摘：论述了抗震设计样板规范的由来和发展，探讨了样板规范的作用和特点，提出了在我国编制和发展样板规范的必要性和途径，且提出了一套较完整的、基于性态设计的工程结构使用功能分类和抗震设计分类的原则、框架和方法。

To analyze the reliability and comparability of different hematology analyzers and to standardize the operation of a contrast test. Methods The carryover rate, precision, total precision and percentage of differences of hematological analyzers were evaluated and differences of items in whole blood were compared by detecting 20 anticoagulant fresh blood samples based on the documents of ICSH and NCCLS. Every 3 blood samples of highvalue, middlevalue and lowvalue were detected by hematological analyzers everyday for 7 days.

方法参照国际血液学标准化委员会及美国临床实验室标准化委员会文件，在仪器正常运行的情况下，评价各比对仪器的携带污染率、精密度和总精密度、差异百分率；分析比较20份抗凝新鲜全血中各项指标在不同血细胞分析仪之间的差异；每日取高、中、低3份新鲜血，在各仪器上平行检测，连续比对7日，观察各项指标与参考样机之间的相关性；分析比较仪器分类与手工分类的相关性；通过直线回归方程计算相对偏差，估计每台仪器各检测指标的试验误差。

It has been put forward that there exists single Ball point and double Ball points on the symmetrical connecting-rod curves of equilateral mechanisms.

从鲍尔点的形成原理出发，分析对称连杆曲线上鲍尔点的产生条件，提出等边机构的对称连杆曲线上有单鲍尔点和双鲍尔点。

The factory affiliated to the Group primarily manufactures multiple-purpose pincers, baking kits, knives, scissors, kitchenware, gardening tools and beauty care kits as well as other hardware tools, the annual production value of which reaches US$ 30 million dollars.

集团所属工厂主要生产多用钳、烤具、刀具、剪刀、厨具、花园工具、美容套等五金产品，年生产总值3000万美元，产品价廉物美、选料上乘、质量保证，深受国内外客户的青睐