检索结果-南通市图书馆

Mining and Integrating Reliable Decision Rules for Imbalanced Cancer Gene Expression Data Sets

在线全文

学校读者我要写书评

暂无评论

Tsinghua Science and technology 2012年第6期17卷 666-673页

作者： Hualong Yu 1 , Jun Ni 2 , Yuanyuan Dan 3 , Sen Xu 4.1. school of Computer Science and engineering, Jiangsu University of Science and technology, Zhenjiang 212003, china 2. Department of Radiology, Carver College of Medicine, The University of Iowa, Iowa City, IA 5224., USA 3. school of Biology and Chemical engineering, Jiangsu University of Science and technology, Zhenjiang 212003, china 4. school of information engineering, yancheng institute of technology, yancheng 2.4051, china school of Computer Science and engineering Jiangsu University of Science and Technology Zhenjiang 212003 China|c|

There have been many skewed cancer gene expression datasets in the post-genomic era. Extraction of differential expression genes or construction of decision rules using these skewed datasets by traditional algorithms will seriously underestimate the performance of the minority class, leading to inaccurate diagnosis in clinical trails. This paper presents a skewed gene selection algorithm that introduces a weighted metric into the gene selection procedure. The extracted genes are paired as decision rules to distinguish both classes, with these decision rules then integrated into an ensemble learning framework by majority voting to recognize test examples; thus avoiding tedious data normalization and classifier construction. The mining and integrating of a few reliable decision rules gave higher or at least comparable classification performance than many traditional class imbalance learning algorithms on four benchmark imbalanced cancer gene expression datasets.

关键词： cancer gene expression data class imbalance paired differential expression genes decision ruleensemble learning majority voting

Generalized Self-Adaptive Genetic Algorithms

在线全文

学校读者我要写书评

暂无评论

International Journal of Minerals,Metallurgy and Materials 2000年第1期14卷 72-75页

作者： Bin Wu Xuyan Tu Jian Wu information engineering school, University of Science and technology Beijing, Beijing 100083,{4. Department of information and Control engineering, Southwest institute of technology, Mianyang 621002,{4. information engineering school University of Science and Technology Beijing Beijing 100083 China Department of information and Control engineering Southwest Institute of Technology Mianyang 621002 China

In order to solve the problem between searching performance and convergence of genetic algorithms, a fast genetic algorithm generalized self-adaptive genetic algorithm (GSAGA) is presented. (1) Evenly distributed initial population is generated. (2) Superior individuals are not broken because of crossover and mutation operation for they are sent to subgeneration directly. (3) High quality im- migrants are introduced according to the condition of the population schema. (4. Crossover and mutation are operated on self-adaptation. Therefore, GSAGA solves the coordination problem between convergence and searching performance. In GSAGA, the searching per- formance and global convergence are greatly improved compared with many existing genetic algorithms. Through simulation, the val- idity of this modified genetic algorithm is proved.

关键词： generalized self-adaptive genetic algorithm initial population immigration fitness function

在线全文

学校读者我要写书评

暂无评论

基于跨证据文本实体关系构建的事实核查研究

中文信息学报 2024年第3期38卷 93-101,112页

作者：贺彦程徐冰朱聪慧哈尔滨工业大学计算机学院黑龙江哈尔滨150001

事实核查是指基于证据文本的虚假信息检测任务,目前已有的研究方法主要是将声明文本与证据文本拼接后输入预训练模型进行分类判断,或者通过单一节点的全连接图进行推理判断。这些方法忽略了证据文本间的远距离语义关联和其包含的噪声干... 详细信息

事实核查是指基于证据文本的虚假信息检测任务,目前已有的研究方法主要是将声明文本与证据文本拼接后输入预训练模型进行分类判断,或者通过单一节点的全连接图进行推理判断。这些方法忽略了证据文本间的远距离语义关联和其包含的噪声干扰。针对以上问题,该文提出了一种基于跨证据文本实体关系的图卷积神经网络模型(C ross-E vidence Entity R elation Reasoning M odel,CERM)。该模型以多个证据文本的实体共现关系为基础,聚合不同实体对象的语义结构信息,同时减小噪声信息干扰,有效提升模型的虚假信息判别能力。实验结果证明,在公开数据集上该文提出的方法在通用评测指标上均优于现有的对比模型,验证了CERM模型在事实核查研究任务上的有效性。

关键词：事实核查图卷积神经网络实体关系

基于改进YOLOv4.Tiny的机械零件目标检测算法

在线全文

学校读者我要写书评

暂无评论

传感技术学报 2024年第2期37卷 303-309页

作者：杜保帅房桐赵燕成赵景波青岛理工大学信息与控制工程学院山东青岛266520

随着智能化生产的普及,机械零件的智能装配技术受到广泛关注,为了改善传统特征提取算法中检测误差大、精度低等问题,以常见机械零件为研究目标,结合深度学习算法中的轻量级网络为基础模型进行优化,使用YOLOv4.Tiny中的CSP-Darknet53作... 详细信息

随着智能化生产的普及,机械零件的智能装配技术受到广泛关注,为了改善传统特征提取算法中检测误差大、精度低等问题,以常见机械零件为研究目标,结合深度学习算法中的轻量级网络为基础模型进行优化,使用YOLOv4.Tiny中的CSP-Darknet53作为特征提取网络,在特征提取网络后添加改进后的MA-RFB模块,引入多分支卷积和空洞卷积加强感受野。并对颈部网络进行改进,选择PANet代替FPN,并添加注意力模块CBAM,形成CM-PANet对零件目标进行多尺度检测,在自制的零件数据集AP达到96.4.%,检测速度达到0.00138 s每样本。相比于原版YOLOv4.Tiny网络AP提高了2.80%,改进后的算法在速度和精度达到了一个平衡,体现了研究的理论和应用价值。

关键词：图像分类轻量级网络深度学习机械零件感受野

Perceptual video hashing robust against geometric distortions

在线全文

学校读者我要写书评

暂无评论

Science china(information Sciences) 2012年第7期55卷 1520-1527页

作者： XIANG ShiJun 1,4.,YANG JianQuan 2 & HUANG JiWu 3 1 school of information Science and technology,Jinan University,Guangzhou 510632,china 2 Shenzhen {4.s of Advanced technology,Chinese Academy of Sciences,Shenzhen 518055,china 3 school of information Science and technology,Sun Yat-sen University,Guangzhou 510275,china 4.State Key Laboratory of information Security,{4. of Software,Chinese Academy of Sciences,Beijing 10004.,china 1. school of information Science and technology Jinan University Guangzhou 510632 China4. State Key Laboratory of Information Security Institute of Software Chinese Academy of Sciences Beijing 100049 China2. Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Shenzhen 518055 China3. School of Information Science and Technology Sun Yat-sen University Guangzhou 510275 China

In this paper,we propose a robust perceptual hashing algorithm by using video luminance histogram in *** underlying robustness principles are based on three main aspects:1) Since the histogram is independent of position of a pixel,the algorithm is resistant to geometric deformations; 2) the hash is extracted from the spatial Gaussian-filtering low-frequency component for those common video processing operations such as noise corruption,low-pass filtering,lossy compression,etc.; 3) a temporal Gaussian-filtering operation is designed so that the hash is resistant to temporal desynchronization operations,such as frame rate change and *** a result,the hash function is robust to common geometric distortions and video processing *** results show that the proposed hashing strategy can provide satisfactory robustness and uniqueness.

关键词： video hashing geometric distortion histogram Gaussian filtering

Analysis of forest backscattering characteristics based on polarization coherence tomography

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Science china(Technological Sciences) 2010年第S1期53卷 166-175页

作者： LUO HuanMin 1,LI XiaoWen 1,2,CHEN ErXue 3,CHENG Jian 4.& CAO ChunXiang 21 institute of Geo-Spatial information Science and technology,University of Electronic Science and technology of{4.,Chengdu 610054.china 2 State Key Laboratory of Remote Sensing Science,Jointly Sponsored by the institute of Remote Sensing Applications of Chinese Academy of Sciences and Beijing Normal University,Beijing 100101,china 3institute of Forest Resources information Technique of Chinese Academy of Forestry,Beijing 100091,china 4.school of Electronic engineering,University of Electronic Science and technology of{4.,Chengdu 610054.china State Key Lab Remote Sensing Sci Beijing 100101 Peoples R China Beijing Normal Univ Beijing 100101 Peoples R China Univ Elect Sci & Technol china Inst Geospatial Informat Sci & Technol Chengdu 610054 Peoples R China Chinese Acad Forestry Inst Forest Resources Informat Tech Beijing 100091 Peoples R China Univ Elect Sci & Technol china Sch Elect Engn Chengdu 610054 Peoples R China

It is difficult to make an inventory of vertical profiles of forest structure parameters in field ***,analysis and understanding of forest backscattering characteristics contribute to estimation and detection of forest vertical structure because of the close relationships between backscattering characteristics and structure *** vertical structure function in the complex interferometric coherence definition,which represents the vertical variation of microwave scattering with the penetration depth at a point in the 2-D radar image and can be used to analyze the forest backscattering characteristics,can be reconstructed from polarization coherence tomography(PCT).Based on PCT,the paper analyzes the forest backscattering characteristics and explores the inherent relationship between the result of PCT and the forest structure parameters from numerical simulation of Random Volume over Ground model(RVoG),Polarimetric SAR interferometry(PolInSAR)simulation of forest scene and PolInSAR data at L-band of the test site ***,the effects of the extinction coefficient and surface-to-volume scattering ratio in RVoG model on vertical backscattering characteristics are analyzed by means of numerical ***,by applying PCT to L-band POLInSAR simulations of forest scene,different variations of vertical backscattering due to different extinction coefficients and the ratios of surface-to-volume scattering resulting from different polarizations,forest types and densities are displayed and *** a concept of relative average backscattering intensity is presented,and the factors which affect its vertical distribution are also *** results show that there is high sensitivity of the vertical distribution of forest relative average backscattering intensity to the polarization,forest type and ***,based on repeat pass DLR E-SAR L-band airborne POLInSAR data,the capability of PCT technology for detection

关键词： backscattering characteristic POLInSAR PCT vertical structure function

A multicast routing algorithm with multiple trees

在线全文

学校读者我要写书评

暂无评论

Journal of Systems engineering and Electronics 2005年第2期16卷 426-432页

作者： Cui Xunxue 1,2 , Gao Wei3 & Fang Hongyu4.. New Star Research institute of Applied technology, Hefei 230031, P. R. china 2. Jiangsu Key Laboratory of Computer information Processing technology,Soochow University, Suzhou 215006, P. R. china 3. Department of Electronic engineering and information Science,University of Science and technology of china, Hefei 230027, P. R. china 4. school of Electronic Science and technology, Anhui University, Hefei 230039, P. R. china New Star Research institute of Applied technology Hefei 230031 P. R. China 2. Jiangsu Key Laboratory of Computer information Processing technology Soochow University Suzhou 215006 P. R. China Department of Electronic engineering and information Science University of Science and Technology of China Hefei 230027 P. R. China school of Electronic Science and technology Anhui University Hefei 230039 P. R. China

Quality of service (QoS) multicast routing has continued to be a very important research topic in the Internet. A method of multicast routing is proposed to simultaneously optimize several parameters based on multiobjective genetic algorithm, after the related work is reviewed. The contribution lies on that the selection process of such routing is treated with multiobjective optimization. Different quality criterions in IP network are taken into account for multicast communications. A set of routing trees is generated to approximate the Pareto front of multicast problem. Multiple trees can be selected from the final set of nondominated solutions, and applied to obtain a good overall link cost and balance traffic distribution according to some simulation results.

关键词： multicast routing quality of service multiobjective optimization genetic algorithms.

The crustal structure under Sanjiang and its dynamic implications:Revealed by seismic reflection/refraction profile between Zhefang and Binchuan,Yunnan

在线全文

学校读者我要写书评

暂无评论

Science china Earth Sciences 2005年第9期48卷 1329-1336页

作者： ZHANG Zhongjie1, BAI Zhiming1, WANG Chunyong2, TENG Jiwen1, Lü Qingtian3, LI Jiliang1, LIU Yifeng1 & LIU Zhenkuan4.1. institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, china 2. institute of Geophysics, china Seismological Bureau, Beijing 100085, china 3. institute of Deposition Resource, Chinese Academy of Geosciences, Beijing 100037, china 4. school of Exploration and information, china University of Geosciences, Beijing 100083, china 1. institute of Geology and Geophysics Chinese Academy of Sciences 100029 Beijing China 2. institute of Geophysics China Seismological Bureau 100085 Beijing China 3. institute of Deposition Resource Chinese Academy of Geosciences 100037 Beijing China 4. school of Exploration and information China University of Geosciences 100083 Beijing China

The fault belts in Sanjiang mainly include Jinshajiang-Honghe fault, Lancangjiang fault and Nujiang fault (called Sanjiang faults) in western Yunnan Province, china. By interpreting the wide-angle seismic reflection/refraction profile between Zhefang and Binchuan, which crosses Tengchong and Baoshan blocks in Dianxi (western Yunnan) tectonic zone, we recon- struct the crustal structure with seismic traveltime tomography for crustal P-wave velocity and the seismic scattering image for crustal seismic reflection structure. In this paper, we firstly present the crustal structure images of P-wave velocity and seismic reflection under the wide-angle seismic profile. These results demonstrate that, the crustal velocity structure and seismic reflec- tion structure along the profile can be divided into 3 segments, and there is an obvious difference of crustal structure among the eastern, the western and the middle segment. Generally, crustal P-wave velocities in the Baoshan segment are 0.1―0.2 km/s slower and seismic reflection am- plitudes from Moho discontinuity are stronger than the other 2 segments. In the studied area, crustal thickness is about 4. km, and shows the thickening tendency from west to east along the profile. Additionally, it can be seen that there is one strong-amplitude seismic reflection event as bright points at the depths of 8―10 km, along the segment of 80―115 km of the profile (south- ward of Tengchong); and seismic reflection wave-field from Moho discontinuity varies obviously along the lateral direction. Finally, we make some discussions on the crustal thickening pattern in the Sanjiang fault belt, structural environment of earthquake development and the contact rela- tionship between the Tengchong block, Banshan block and Luxi trough.

关键词： Sanjiang area wide-angle seismic profile crustal structure.

Anticoagulation property and security of artificial heart valve material

在线全文

学校读者我要写书评

暂无评论

Science china(Technological Sciences) 2005年第2期48卷 145-162页

作者： WANG Xianghui1, , LIU Xianghuai1,22, 3, ZHANG Feng2, LI Changrong2, YU Liujiang2, ZHENG Zhihong2, WANG Xi2, JIANG Zhenbin4. CHEN Anqing4.& XU Jingfang3 1. Department of Physics, East china Normal University, Shanghai 200062, china 2. Ion Beam Lab, Shanghai institute of Mini-system and information, Chinese Academy of Sciences,Shanghai 200050, china 3. Department of Electronics, school of information Science and technology, East china Normal Univer- sity, Shanghai 200062, china 4. Shanghai Cardiovascular institute, Zhongshan Hospital, Shanghai 200032, china 1. Department of Physics East China Normal University 200062 Shanghai China 2. Ion Beam Lab Shanghai Institute of Mini-system and information Chinese Academy of Sciences 200050 Shanghai China 3. Department of Electronics School of Information Science and Technology East China Normal University 200062 Shanghai China 4. Shanghai Cardiovascular institute Zhongshan Hospital 200032 Shanghai China

Heart valve diseases threaten human health. One reliable way to save lives of such patients is to replace the pathologically changed heart valves by artificial ones. Over 2 million patients have received LTI-carbon heart valve’s implantation. However, the thrombosis after the implantation is one of the difficulties that need to be solved. In order to improve the blood compatibility and security of mechanical heart valves, ion beam technology was used to modify the surface properties of the materials. The investigation results have been summarized in this paper.

关键词： artificial heart valve, ion implantation, plasma immersed ion implantation, ion beam enhanced deposition.