抽取的

With the deficit in the information processing on facial emotion perception of schizophrenic patients, the neurocognitive training program is based mainly on the Memory Processing Model, Treisman's Attenuation Model, signs of the eye movement, Ellis's Information Processing Model of Face Recognition, Zhuo's Morphological-Tuning Model, Wang's Two-Stage Face Recognition Model, and Anderson's Adaptive Control of Thought. The contents of the training contain self-instruction, feature abstraction, mnemonics, and positive reinforcement.

本研究以记忆历程模式为主轴，参考减弱模式的感官登录与减弱控制机转、眼球凝视现象、脸部辨识讯息历程模式、双阶段脸孔辨识模式的型态侦测模块、二阶段脸孔辨识模式的脸孔与表情表征、思考适应控制模式的陈述与程序知识等部分，并举证精神分裂病患在表情讯息处理历程各阶段－注意、编码、储存、提取的缺损现象，以自我引导、特征抽取、记忆增强、立即回馈等原则，设计神经认知行为取向的脸部表情辨认训练方案。

S.From all the patients with MB-MCA and whose MCA were encased by the myocardium completely,30 patients were choosed randomly,their datasets were reconstructed seriatim by every 5%R-R interval,the whole cardiac cycle were reconstructed by 20 phases,the diameter of the mural coronary arteries of every phase was reviewed and measured,the phases were found out where the diameters were maximal and minimal,the systolic compression rate was calculated,Pearson correlation analysis was used to analyze the relation between length,depth,and degree of systolic compression of the mural coronary artery.

s表示。在所有存在MB-MCA并且MCA被心肌完全包饶的病例图像中，随机抽取30例图像的数据，以每隔5%R-R间期为一个重建时相，对每个病例0.100%整个心动周期分20个期相进行重建，观察并测量MCA在整个心动周期中管径的变化，探索管径最大和最小时的时相显示规律，计算MCA最大狭窄程度，应用Pearson相关统计分析MCA狭窄程度与MB的长度及厚度之间的相关关系。

As an important component among information acquisition methods, the requirement for high quality auto-summarization system also becomes more urgency. To fulfill this requirement, an adapted Dotplot methods based on word sense quantization model is firstly provided to address the problem of subtopic segmentation. Then a multi-knowledge-sources-integrating model is constructed, which combines the results of rhetorical structure analyse, text content structure analyse and subtopic segmentation together and provides clues for abstract sentences extracting.

第四，互联网的普及增加了人们对信息获取速度、准确率的要求，也推动了人们对自动文摘系统的需求，为了进一步改进现有文摘技术，本文首先研究了基于语义量化模型的改进Dotplot隐式章节划分算法，然后通过加权有向图来构造了一个多知识源融合模型，模型能够融合文本修辞结构分析、内容层次结构分析以及隐式章节划分等多种知识源的分析结果，并以此为基础建立文摘语句的抽取算法。

In order to avoid being unable to select those residences with unlisted telephone numbers, we first took a systematic random sample of telephone numbers, and then added one to the last digit.

首先先依等比例的原则计算各县市应抽取出的样本数，其次依等距抽样的原则抽出电话号码。为避免未登录电话号码的用户无法被抽中，遂在抽出的电话号码尾数加 1，作为受访的样本户。

In addition, the concept of scene centriod is introduced to represent video scene by averaging of all key frames inside a group of shots in the given video scene, which makes an audacious attempt for high-level video unit extraction and representation;4 According to the given sample video, shot and scene-based video retrieval methods are discussed respectively, and the evaluation pricinples are also discussed.

进而，讨论了由场景中一组镜头的关键帧的质心表示场景关键帧，即场景质心的方法，从而为抽取高层的视频数据单元做出了大胆的尝试； d 从给定样本视频出发，分别讨论了基于镜头和场景的视频检索方法，并分析了检索性能评价的准则。

This dissertation brings forward and realizes the multilevel classifiable method which is based on characters coding. Above all, this method realizes the first-grade classification by extracting enough effective characters from characters and coding them, to the others which couldn't be recognized by the first-grade classification, the method will adopt the second-grade classification using template matching to recognize these characters.

本文使用了基于特征编码的多级分类识别方法，通过给字符抽取足够多的有效的特征并给特征编码实现第一级分类，对于第一级分类后仍不能区分的字符，再进入第二级分类用模板匹配的方法最终达到区分的目的，这种方法的重点在第一级分类阶段。

This dissertation brings forward and realizes the multilevel classifiable method which is based on characters coding. Above all, this method realizes the first-grade classification by extracting enough effective characters from characters and coding them, to the others which coundn t be recognized by the first-grade classification, the method will adopt the second-grade classification using template matching to recognize these characters.

本文提出并实现了基于特征编码的多级分类识别方法，通过给字符抽取足够多的有效的特征并给特征编码实现第一级分类，对于第一级分类后仍不能区分的字符，再进入第二级分类用模板匹配的方法最终达到区分的目的，这种方法的重点在第一级分类阶段。

Objective Suction body fat in the eyelids filled with depression,enophthalmos used to repair the eyeliddeformity Methods Ordinary syringe suction side of the thigh,from the requirements of eyelid fat particles injectedenophthalmos Department.

目的：通过抽吸自体的脂肪颗粒填充于眼睑的凹陷处，用来修复眼睑的内陷畸形。方法：用普通的注射器抽吸大腿的前侧面，抽取需要量的脂肪颗粒注射于眼睑内陷处。

Methods A randomized cluster sampling was performed to obtain a representative sample of population in Xinjiang Hefeng pastoral area.Blood wag taken from 632 individuals aged 30 and older for measurements of total cholesterol,triglyceride,hish density lipoprotein cholesterolby automatic biochemieal analyzer.The serum levels of low density lipoprotein cholesterolwas calculated according to the formula.

采用整群随机抽样的方法选取新疆和丰县牧区年龄≥30岁的牧民632人为调查对象，其中哈萨克族325人，蒙古族307人；抽取空腹12 h静脉m 3ml，采用日立7600全自动生化分析仪测定血浆总胆固醇、甘油三酯、高密度脂蛋白胆固醇的浓度，并根据公式计算出LDL-C的浓度，对资料进行汇总，并采用t检验、方差分析或协方差分析的方法观察两民族间血浆LDL-C水平的差异，并进一步采用多元逐步同归分析的统计方法研究其影响因素。

Nowadays with the wide use of technology of Internet, the problem of "adequate data and scarcity knowledge" has become increasingly distinct. Knowledge discovery study how to gain valuable knowledge and information intellectively and automatically from a mass of data, so it has been a very active studying field in artificial intelligent.

随着Internet技术的日益普及，&丰富的数据与贫乏的知识&问题变得日渐突出，而知识发现正是研究如何从大量的数据中智能地、自动地抽取出有价值的知识和信息，因而成为当前人工智能中非常活跃的研究领域。

She gently rebuff ed him, but agreed that they could be friends

她婉言拒绝了，但同意作为朋友相处。

If in the penal farm, you were sure to be criticized.

要是在劳改农场，你等着挨绳子吧！