检索结果-内蒙古大学图书馆

Using Parallel Partitioning Strategy to Create Diversity for Ensemble Learning

Using Parallel Partitioning Strategy to Create Diversity for...

2009 2nd IEEE International Conference on computer science and Information technology(第二届计算机科学与信息技术国际会议 ICCSIT2009)

作者： Yi-Min Wen Yao-Nan Wang Wen-Hua Liu College of Electrical and Information Engineering Hunan UniversityChangshaHunanChina Department College of Electrical and Information Engineering Hunan UniversityChangshaHunanChina Department of Computer Science and Technology Hunan Institute of TechnologyHengyangHunanChina

Divide-and-conquer principle is a fashionable strategy to handle large-scale classification problems. However, many works have revealed that generalization ability is decreased by partitioning training set in most cases, because partitioning training set can lead to losing classification information. Aiming to handle this problem, an ensemble learning algorithm was proposed. It used many sets of parallel hyperplanes to partition training set on which each base classifier was trained by the SVM modular network algorithm and all these base classifiers were combined by majority voting strategy when testing. The experimental results on 4 classification problems illustrate that ensemble learning can effectively reduce the descent of generalization ability for the reason of increasing classifier’s diversity.

关键词： Partitioning algorithms Testing Machine learning Machine learning algorithms Clustering algorithms Large-scale systems Support vector machines Support vector machine classification Industrial training Voting

来源：评论

学校读者我要写书评

暂无评论

Principal Component Analysis for Prediction of O-Linked Glycosylation Sites in Protein by Multi-Layered Neural Networks

Principal Component Analysis for Prediction of O-Linked Glyc...

引用

International Conference on Intelligent Information Hiding and Multimedia Signal Processing (IIHMSP)

作者： Chu-Zheng Wang Xian-Hua Han Masahiro Ito Ikuko Nishikawa Yen-Wei Chen College of Computer Science Central South University of Forestry and Technology ChangSha Hunan China College of Information Science and Engineering Ritsumeikan University Shiga Japan

Glycosylation is one of the common post-translation modification of protein in eukaryotic cells. Conventional neural network methods have been applied to predict glycosylation sites in protein sequence and the prediction accuracy is dependent on the dimension of feature vector (length of protein sequence). Though the prediction accuracy can be improved by increasing the length of protein sequence, it is time-consuming. In this paper, we propose a novel approach which combines PCA with a multilayer neural network for efficient, and accurate prediction of O-glycosylation sites in protein. PCA is first used to extract main basis (subspace) of the protein sequence. The lower-dimensional projections on the subspace are used as features instead of the higher-dimensional protein sequences. Compared with conventional method (without PCA), our proposed method can significantly reduce the larger computation cost, while can keep the higher prediction accuracy.

关键词： Principal component analysis Neural networks Multi-layer neural network Amino acids Protein sequence Accuracy Educational institutions Computational efficiency computer networks Intelligent networks

来源：评论

学校读者我要写书评

暂无评论

Connected dominating sets in wireless sensor networks

Connected dominating sets in wireless sensor networks

引用

5th International Conference on Wireless Communications, Networking and Mobile Computing, WiCOM 2009

作者： Zhang, Lili Li, Chengming Mao, Yingchi Xu, Lizhong Wang, Huibin College of Computer Science and Inforrmation Engineering Hohai University Nanjing 210098 China Nanjing University State Key Laboratory for Novel Software Technology Nanjing 210098 China

ISBN: (纸本)9781424436934

Since there is no fixed infrastructure or centralized management in wireless sensor networks, a Connected Dominating Set(CDS) has been proposed to serve as a virtual backbone. The CDS of a graph representing a network has a significant impact on the efficient design of routing protocols in wireless sensor networks. This problem has been studied extensively, in which all nodes have the same transmission ranges and the size of the dominating sets are considered. In this paper, by the method of constructing the steiner tree, we model a network as an undirected graph and present the CDS which has smaller size. The performance ratio of these algorithms is constant. Furthermore, we show a size relationship between a maximal independent set and a CDS as well as a bound of the maximum number of independent neighbors of a node. The theoretical analysis and simulation results are also presented to verify our approaches. ©2009 IEEE.

关键词： Sensor nodes

来源：评论

学校读者我要写书评

暂无评论

Color Image Stereo Correspondence Algorithm for Obstacle Detection

Color Image Stereo Correspondence Algorithm for Obstacle Det...

引用

Chinese Conference on Pattern Recognition (CCPR)

作者： Long-yuan Guo Yong-quan Xia Department of Computer science Hunan Institute of Science and Technology Yueyang China College of Computer and Communication Engineering Zhengzhou University of Light Industry Zhengzhou China

Obstacle detection is a crucial problem of autonomous vehicle navigation. The paper presents color image stereo correspondence algorithm for obstacle detection. Firstly, the color images are segmented in HSV color space. And the regions of interest are extracted from the images; Then, the rectangular boundary of the interest regions are calculated; Obstacle position finally are obtained using stereo correspondence method. Segmenting and extracting of obstacle object is the key during this process. The image in HSV color space in the paper has compact information structure. And every component of HSV color space is more independence. Image segment is more accurately with these characteristics. The result of experiment indicates that it is effect for detection of road, vehicles and house and suppression of natural background.

关键词： Color Image segmentation Vehicle detection Space technology Data mining computer science Educational institutions Automotive engineering Communication industry computer industry

来源：评论

学校读者我要写书评

暂无评论

Effects of polypropylene-g-dibutyl maleate on mechanical and rheological properties of PP/PA6 blends

引用

E-Polymers 2009年

作者： Li, Duxin Ou, Baoli Jia, Demin State Key Laboratory of Powder Metallurgy Central South University Changsha 410083 China School of Chemistry and Chemical Engineering Hunan University of Science and Technology Xiangtan 411201 China College of Materials Science and Engineering South China University of Technology Guangzhou 510640 China

The effects of polypropylene-grafted-dibutyl maleate (PP-g-DBM) used as compatibilizer on mechanical, rheological and morphological properties of polypropylene (PP) and nylon 6 (PA6) blends was systematically investigated in this paper. The results of Molau test, solvent extraction and differential scanning calorimetry (DSC) indicated the formation of PP-g-PA6 in vicinity of interfaces during melting extrusion. Owing to the reaction between the reactive groups of PP-g-DBM and amine (-NH2) end groups of PA6, the tensile and impact strength of the PP-g-DBM compatibilized PP/PA6 blends were much higher than that of the uncompatibilized PP/PA6 blends. Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) results indicated that the domain sizes of the dispersed phase in the compatibilized PP/PA6 blends decreased and the interfaces become more indistinct, which indicated a clear compatibiliting effect was induced by the PP-g-DBM in the immiscible PP/PA6 blends, i.e. the PP-g-DBM was an effective compatibilizer for the PP/PA6 blends. In addition, the rheological measurements showed the PP-g-DBM compatibilized PP/PA6 blends possessed higher pseudo plasticity, melt viscosity and flow activation energy.

关键词： Polypropylenes

来源：评论

学校读者我要写书评

暂无评论

The research of power grid data integration and sharing platform based on SOA

The research of power grid data integration and sharing plat...

引用

2nd International Workshop on computer science and engineering, WCSE 2009

作者： Wu, Kehe Shi, Qianhui Duan, Cheng Yin, Haiyani Department of Computer Science and Technology North China Electric Power University Beijing China College of Electrical and Electronic Engineering North China Electric Power University Beijing China

ISBN: (纸本)9780769538815

Under the circumstance that electricity grid is becoming more and more complex and enormous, management information systems should be able to provide a complete and comprehensive information, which puts forward the need to integrate the various data in the grid enterprises. Followed by this need, combined with service-oriented architecture (SOA) and Web Service technology, this paper presents a service-oriented platform for data integration and sharing. Besides that it also discusses the concrete realization of the platform process. The practical application proves that the SOA-based grid platform for data integration and sharing shows the good scalability and practicality. © 2009 IEEE.

关键词： Web services

来源：评论

学校读者我要写书评

暂无评论

Multi-objective ant colony optimization biclustering of microarray data

Multi-objective ant colony optimization biclustering of micr...

引用

IEEE International Conference on Granular Computing (GRC)

作者： Junwan Liu Zhoujun Li Xiaohua Hu Yiming Chen School of Computer and Information Engineering Central South University of Forestry and Technology China School of Computer Science & Engineering Beihang University China College of Information Science and Technology Drexel University USA School of Information Science and Technology Hunan Agricultural University China

Latest microarray technique can measure the expression levels of thousands of genes under a set of conditions, and generates some large-scale microarray datasets. Biclustering can perform clustering of rows and columns of those dataset simultaneously, allowing the mining of additional information from microarray datasets which is important in bioinformatics research and biomedical applications. Since the biclustering problem is combinatorial, and multi-objective ant optimization systems present several advantages during dealing with this kind of problem. This paper proposes a novel multi-objective ant colony optimization biclustering algorithm to mine biclusters from microarray dataset. Experimental results on real dataset show that our approach can find significant biclusters of high quality.

关键词： Ant colony optimization Gene expression Bioinformatics Diseases Clustering algorithms Data engineering Forestry Information science Biomedical measurements Large-scale systems

来源：评论

学校读者我要写书评

暂无评论

Predicting gene function with positive and unlabeled examples

Predicting gene function with positive and unlabeled example...

引用

IEEE International Conference on Granular Computing (GRC)

作者： Yiming Chen Zhoujun Li Xiaohua Hu Hongxiang Diao Junwan Liu National University of Defense Technology Changsha Hunan China School of Information Science and Technology Hunan Agricultural University Changsha Hunan China School of Computer Science and Engineering Beihang University Beijing China College of Information Science and Technology Drexel University Philadelphia PA USA

Predicting gene function is usually formulated as binary classification problem. However, we only know which gene has some function while we are not sure that it doesn't belong to a function class, which means that only positive examples are given. Therefore, selecting a good training example set becomes a key step. In this paper, we cluster the genes on integrated weighted graph by generalizing the cluster coefficient of unweighted graph to weighted one, and identify the reliable negative samples based on distance between a gene and centroid of positive clusters. Then, the tri-training algorithm is used to learn three classifiers from labeled and unlabeled examples to predict the gene function by combining three prediction result. The experiment results show that our approach outperforms several classic prediction methods.

关键词： Information science computer science Agricultural engineering Educational institutions Clustering algorithms Prediction methods Large-scale systems Genomics Bioinformatics Proteins

来源：评论

学校读者我要写书评

暂无评论

Genetic sampling in eigenspace for 3D caricature synthesis

Genetic sampling in eigenspace for 3D caricature synthesis

引用

International Conference on Information, Communications and Signal Processing

作者： Xingyu Gao Yiqiang Chen Jingye Zhou Junfa Liu Chunyan Miao Zhenyu Chen College of Information Engineering Xiangtan University Xiangtan Hunan China Institute of Computing Technology Chinese Academy and Sciences Beijing China School of Computer Engineering Nanyang Technological University Singapore

ISBN: (纸本)9781424446568

Caricature synthesis is becoming one of the most challenging research problems nowadays, especially for 3D caricature. In this paper, we present a novel approach for 3D caricature synthesis adopting genetic sampling in eigenspace. Our approach aims to generate 3D caricature which has not only exaggerated effect but also facial feature of real face. Firstly, we build 3D caricature components eigenspaces by Principal Component Analysis (PCA). Moreover, genetic algorithm (GA) is used to sample optimal projections in the eigenspaces for 3D real face components, and then we reconstruct 3D caricature components by the projections. At last, we refine the 3D caricature using PCA-based sampling approach and Kriging interpolation. The experiment results show that our approach can effectively synthesize interesting and harmonious 3D caricature.

关键词： Sampling methods Principal component analysis Spatial databases Facial features Genetic engineering Interpolation Educational institutions computers Magnetic heads Deformable models

来源：评论

学校读者我要写书评

暂无评论

Rough set methods based on discernibility relationship in IIS

Rough set methods based on discernibility relationship in II...

引用

IEEE International Conference on Granular Computing (GRC)

作者： Yanlin Li Lihua Wei School of Electrical Engineering Hunan University of Technology Zhuzhou China School of Computer Science and Technology Nanjing University of Science and Technology of PLA Nanjing China College of Information Science & Technology Drexel University PA USA

Classical rough set theory based on indiscernibility relationship rarely concerns discernibility relationships. However discernibility relationship deserves more attention because of its higher efficiency. In this paper a new kind of discernibility relationship is defined for studying incomplete information system. To demonstrate the validity of discernibility relationship, first we give a precise definition of discernibility relationship before the corresponding lower and upper approximations can be computed. Next the major properties of the lower and upper approximations can be obtained and proved. Finally the rough set model based on discernibility relationship is so easily constructed.

关键词： Set theory Information systems Information analysis Data analysis Information retrieval Physics

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：