检索结果-内蒙古大学图书馆

2009 International Conference on Machine Learning and Cybernetics(2009机器学习与控制论国际会议)

作者： XUE-FEI BAI WEN-JIAN WANG School of Computer and Information Technology Shanxi University Key Laboratory of Computational Intelligence and Chinese Information Processing of Ministry of Education Taiyuan 030006 China

This paper proposes a novel method for facial expression recognition based on neural network ensemble. The facial expression features are extracted firstly through multi-expression eigenspace analysis, and then several neural networks are trained each with an eigenspace of different expressions respectively. At last their training results are aggregated as inputs of the ensemble classifier, which will provide not only the final recognition results but also the estimated expression information. Simulation results on JAFEE dataset show that the recognition accuracy of the proposed approach is better than that of the best individual neural network.

关键词： Facial ezpression recognition Neural network ensemble Two-dimension principal component analysis

来源：评论

学校读者我要写书评

暂无评论

An attribute reduction approach and its accelerated version for hybrid data

An attribute reduction approach and its accelerated version ...

引用

IEEE International Conference on Cognitive Informatics

作者： Wei Wei Jiye Liang Yuhua Qian Feng Wang School of Computer and Information Technology Shanxi University Taiyuan Shanxi China Chinese Information Processing of Ministry of Education Key Laboratory of Computational Intelligence Taiyuan Shanxi China

In practical issues, categorical data and numerical data usually coexist, and a unified data reduction technique for hybrid data is desirable. In this paper, an information measure is proposed for computing the discernibility power of a categorical or numeric attribute. Based on the measure, a uniform definition of significance of attributes with categorical values and numerical values is proposed. Furthermore, an algorithm to obtain an attribute reduct from hybrid data is presented, and one of its accelerated version is also constructed. Experiments show that these two algorithms can get the same reducts, and the classification accuracies of reduced datasets are similar with the ones using Hu's algorithm. However, the accelerated version consumes much less time than the original one and Hu's algorithm do.

关键词： Acceleration Information systems Power measurement Set theory Entropy Fuzzy set theory Argon Fuzzy sets Laboratories computational intelligence

来源：评论

学校读者我要写书评

暂无评论

A NOVEL LEARNING MODEL-KERNEL GRANULAR SUPPORT VECTOR MACHINE

A NOVEL LEARNING MODEL-KERNEL GRANULAR SUPPORT VECTOR MACHIN...

引用

2009 International Conference on Machine Learning and Cybernetics(2009机器学习与控制论国际会议)

作者： HU-SHENG GUO WEN-JIAN WANG CHANG-QIAN MEN School of Computer and Information Technology Shanxi University Taiyuan 030006 China Key Laboratory of Computational Intelligence and Chinese Information Processing of Ministry of Education Taiyuan 030006 China

This paper presents a novel machine learning model-Kernel Granular Support Vector Machine (KGSVM), which combines traditional support vector machine (SVM) with granular computing theory. By dividing granules and replacing with them in kernel space, the datasets can be reduced effectively without changing data distribution. And then the generalization performance and training efficiency of SVM can be improved. Simulation results on UCI datasets demonstrate that KGSVM is highly scalable for large datasets and very effective in terms of classification.

关键词： Kernel granular support vector machine Support vector machine Kernel space Granules

来源：评论

学校读者我要写书评

暂无评论

Accelerating Incomplete Feature Selection

Accelerating Incomplete Feature Selection

引用

2009 International Conference on Machine Learning and Cybernetics(2009机器学习与控制论国际会议)

作者： Yuhua Qian Jiye Liang Wei Wei Key Laboratory of Computational Intelligence and Chinese Information Processing of Ministry of Education Taiyuan 030006 Shanxi China School of Computer and Information Technology Shanxi University Taiyuan 030006 Shanxi China

Feature selection from incomplete data aims to retain the discriminatory power of original features in rough set theory. Many feature selection algorithms are computationally time-consuming. To overcome this drawback, we introduce a theoretic framework based on rough set theory, called positive approximation, which can be used to accelerate a heuristic process of feature selection from incomplete data. Based on the proposed accelerator, a general feature selection algorithm is designed. Through the use of the accelerator, several representative heuristic feature selection algorithms in rough set theory have been enhanced. Experiments show that these modified algorithms outperform their original counterparts.

关键词： Feature selection Rough set theory Incomplete information systems Positive approzimation Granular computing

来源：评论

学校读者我要写书评

暂无评论

Boid particle swarm optimisation

引用

International Journal of Innovative Computing and Applications 2009年第2期2卷 78-85页

作者： Cui, Zhihua Shi, Zhongzhi Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences No. 6 Kexueyuan South Rd. Zhongguancun Haidian District Beijing 100190 China Complex System and Computational Intelligence Laboratory Taiyuan University of Science and Technology No. 66 Waliu Rd. Wanbailin District Taiyuan Shanxi 030024 China

Particle swarm optimisation (PSO) is a novel population-based stochastic optimisation algorithm inspired by the Reynolds' boid model. The original biological background of boid obeys three basic simple steering rules: separation, alignment and cohesion. However, to promote a simple update equation, none of these rules of boid model is employed by PSO methodology. Due to the weakness of biological background of PSO, in this paper, a new variant of PSO, boid particle swarm optimisation (BPSO), is designed in which cohesion rule and alignment rule are both employed to improve the performance. In BPSO, each particle has two motions: divergent motion and convergent motion. For divergent motion, each particle adjusts its moving direction according to the alignment direction and the cohesion direction, as well as in convergent motion, the original update equation of the standard version of PSO is used. To make a motion transition, a threshold is introduced to make the divergent motion is employed in the first period, whereas the convergent motion is used in the final stage. To testify the efficiency, several unconstrained benchmarks are used to compare. Simulation results show the proposed variant is more effective and efficient than other two variants of PSO when solving multi-modal high-dimensional numerical problems. Copyright © 2009 Inderscience Enterprises Ltd.

关键词： Alignment

来源：评论

学校读者我要写书评

暂无评论

Protein surface Modeling using active contour model 1

引用

5th International Conference on Geometric Modeling and processing (GMP 2008)

作者： Xiang, Junping Hu, Maolin Key Laboratory of Intelligent Computing & Signal Processing Ministry of Education and School of Mathematics and Computational Science Anhui University China

ISBN: (数字)9783540792468

ISBN: (纸本)9783540792451

A PDE (partial differential equation) based method, 3D active contour model, is presented to model the surface of protein structure. Instead of generating a single molecular surface, we create a series of surfaces associated with the atomic energy inside the protein, which describe different resolutions of molecular surface. Our results indicate that the surfaces we generated are suitable for shape analysis and visualization. So, when the solvent-accessible surface is not enough to represent the features of protein structure, the evolution surface sequence may be an alternative choice. Besides, if the initial surface is smooth enough, the generated surfaces will preserve this property partly, because the evolution of the surfaces is controlled by the PDE.

关键词： chan-Vese method solvent-accessible surface evolution partial differential equation finite difference atomic energy

来源：评论

学校读者我要写书评

暂无评论

Information granules and entropy theory in information systems

引用

Science in China(Series F) 2008年第10期51卷 1427-1444页

作者： LIANG JiYe & QIAN YuHua key laboratory of computational intelligence and Chinese Information processing,Ministry of Education School of Computer & Information Technology,Shanxi University,Taiyuan 030006,China 1. Key Laboratory of Computational Intelligence and Chinese Information Processing Ministry of Education School of Computer & Information Technology Shanxi University Taiyuan 030006 China

Information granulation and entropy theory are two main approaches to research uncertainty of an information system, which have been widely applied in many practical issues. In this paper, the characterizations and representations of information granules under various binary relations are investigated in information systems, an axiom definition of information granulation is presented, and some existing definitions of information granulation become its special forms. Entropy theory in information systems is further developed and the granulation monotonicity of each of them is proved. Moreover, the complement relationship between information granulation and entropy is established. This investigation unifies the results of measures for uncertainties in complete information systems and incomplete information systems.

关键词： information systems information granule information granulation entropy rough set

来源：评论

学校读者我要写书评

暂无评论

Consistency and fuzziness in ordered decision tables

引用

3rd International Conference on Rough Sets and Knowledge Technology, RSKT 2008

作者： Qian, Yuhua Liang, Jiye Wei, Wei Wang, Feng Key Laboratory of Computational Intelligence and Chinese Information Processing Ministry of Education Taiyuan Shanxi 030006 China

ISBN: (纸本)3540797203

In this paper, we focus on how to measure the consistency of an ordered decision table and the fuzziness of an ordered rough set and an ordered rough classification in the context of ordered information systems. The membership function of an object is defined through using the dominance class including itself. Based on the membership function, we introduce a consistency measure to assess the consistency of an ordered decision table and define two fuzziness measures to compute the fuzziness of an ordered rough set and an ordered rough classification. Several examples are employed to illustrate their mechanisms as well. These results will be helpful for understanding the uncertainty in ordered information systems and ordered decision tables. © 2008 Springer-Verlag Berlin Heidelberg.

关键词： Decision tables

来源：评论

学校读者我要写书评

暂无评论

Construction method of kernel function based on time correlation

引用

Journal of computational Information Systems 2008年第1期4卷 317-322页

作者： Wang, Ping Wang, Wenjian Zhang, Guisheng Key Laboratory of Computational Intelligence and Chinese Information Processing School of Computer and Information Technology Shanxi University Taiyuan 030006 China Shanxi Police Academy Taiyuan 030021 China

One key point in atmospheric and environmental research today is forecasting of air quality because the health is impacted by the pollutants existing in urban air. Support vector machine (SVM) has been used in air quality prediction as a new learning method developed in recent years. The selection of kernel function is one of important branches in SVM researches. This paper presents a new kernel function based on time correlation for time series data, which incorporates the cyclical feature of time series into SVM. Simulation experiments demonstrate the presented kernel function can help to improve the fitting effect and obtain better generalization performance of SVR.

关键词： Air quality

来源：评论

学校读者我要写书评

暂无评论

Comparisons of protein structure alignment methods: Rigid and flexible, sequential and non-sequential

Comparisons of protein structure alignment methods: Rigid an...

引用

2nd International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2008

作者： Xiang, Junping Hu, Maolin Key Laboratory of Intelligent Computing and Signal Processing School of Mathematics and Computational Science Anhui University Hefei Anhui China

ISBN: (纸本)9781424417483

We try to evaluate the sequential and non-sequential and flexible structural alignment methods on SCOP 1.71. Firstly, we compare the flexible method with rigid methods and compare the sequence order dependent methods with sequence order independent methods by two typical cases. Secondly, the performances of the above methods are evaluated using ten protein pairs which are considered as "hard to detect" similarities. Thirdly, we evaluate the methods by comparing the ROC curves based on their native score and geometric measure Qscore. Then we compare the methods directly using geometric match measure Qscore, which also allows the creation of a 'Best-of-All' method. The main conclusions are: (1) we find that the best alignments found by sequential methods are more consistent with the SCOP classification;(2) we show that ROC curves are of limited value and that their ranking of the methods is not consistent with the ranking implied by the quality of the alignments the methods find;(3) we also find that the alignments of flexible method and non-sequential methods are better than the corresponding alignments of rigid sequential method from the viewpoint of geometric superposition, and flexible method and non-sequential methods can find more cross-fold similarities. © 2008 IEEE.

关键词： Proteins

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：