检索结果-内蒙古大学图书馆

2006 International Conference on Machine Learning and Cybernetics(IEEE第五届机器学习与控制论坛)

作者： GUO-LIANG ZHOU YUN-JIE PANG ZHEN-HUA ZHANG YONG QUAN WEN-HUI LI Computer Graphics Laboratory College of Computer Science and Technology Jilin University Qianwei Key Laboratory for Symbolic Computation and Knowledge Engineering of Ministry of Education

In this paper a new point-based rendering method for ray tracing is presented. An oriented spherical patch that passes a surface point is used to approximate the vicinity of that *** this paper the spherical patch together with the surface point is called "Spherical Patch Point" (SPP). Due to property of curvature, SPP can achieve similar visual quality compared with previous methods with much fewer points. This paper defines new point attributes for the purpose of efficiently locating the intersection between incoming ray and ***, an algorithm of intersecting a ray with point geometry is proposed. The algorithm can achieve a higher rendering speed in comparison with previous methods. The presented technique deals well with shadow, reflection and refraction.

关键词： Point-based rendering Curvature Spherical patch Radius Ray tracing

来源：评论

学校读者我要写书评

暂无评论

A Dynamic Growing Neural Network for Supervised or Unsupervised Learning

A Dynamic Growing Neural Network for Supervised or Unsupervi...

引用

World Congress on Intelligent Control and Automation (WCICA)

作者： Daxin Tian Yanheng Liu Da Wei College of Computer Science and Technology Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun Jilin China

A dynamic growing neural network (DGNN) for supervised learning of pattern recognition or unsupervised learning of clustering is presented. The main ideas included in DGNN are growing, resonance, and post-prune. DGNN is called dynamic growing because it is based on the Hebbian learning rule and adds new neurons under certain conditions. When DGNN performs supervised learning, resonance will happen if the winner can't match the training example; this rule combines the ART/ARTMAP neural network and WTA learning rule. When DGNN performs unsupervised learning, post-prune is carried out to prevent over fitting the training data just like decision tree learning. DGNN's prune rule is based on the distance threshold. DGNN has some advantages: learning not only is stable because it grows under certain conditions; but also it is faster than back-propagation rules and favorable learned predictive accuracy in small, noisy, online or offline data sets. Three classes of simulations are performed on the primary benchmarks: circle-in-the-square and two-spirals-apart benchmarks are used to check DGNN's supervised learning and compare it with ARTMAP and BP neural networks; DGNN's unsupervised learning ability is checked on UCI Machine Learning Archive's Synthetic Control Chart Time Series data set

关键词： Neural networks Unsupervised learning Supervised learning Resonance Pattern recognition Hebbian theory Neurons Subspace constraints Training data Decision trees

来源：评论

学校读者我要写书评

暂无评论

AN EXTENDED RCSP GENERATING MODEL AND ITS APPLICATION

AN EXTENDED RCSP GENERATING MODEL AND ITS APPLICATION

引用

2006 International Conference on Machine Learning and Cybernetics(IEEE第五届机器学习与控制论坛)

作者： JIAN GAO YONG-GANG ZHANG JI-GUI SUN College of Computer Science and Technology Jilin University Changchun 130012 China Key Laboratory for Symbolic Computation and Knowledge Engineering of Ministry of Education Changchu

In recently years there has been plenty of interest in Random Constraint Satisfaction Problem, both from an experimental and a theoretical point of view. In this paper we study and analyze the four popular problem instance generating models, and present the extended model B+ based on the most used model B, which has the different domains and constraint tightness meeting some probability distribution function. In the subsequent section we give the relation matrix version of backtracking integrated forward checking algorithms, and introduce the implementation of instances generator and solver based on the new model. Finally we show the experiment results and conclude the paper, point that our extended model B+ has the common phase transition region with the transitional models and it has the advantage of being suited to the testing of heuristic based constraint solving algorithms, such as variables selection heuristic algorithms.

关键词： Constraint programming RCSP constraint solving generating model testing

来源：评论

学校读者我要写书评

暂无评论

STUDIES ON SOME DETAILS OF ALGORITHM IGRS

STUDIES ON SOME DETAILS OF ALGORITHM IGRS

引用

2006 International Conference on Machine Learning and Cybernetics(IEEE第五届机器学习与控制论坛)

作者： CHUN-XIAO FU SHU-YANG SUN CHENG-MIN SUN DA-YOU LIU College of Computer Science and Technology Jilin University Changchun 130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin Uni

In Containing Order Rough Set Methodology (CORS),terminologies on rules or rules set, such as robust, minimality,completeness, mutuality degree, and conflict are discussed. The rules generation algorithm IGRs is given and the details of algorithm IGRs are studied. Heuristic knowledge, which is mutuality degree of a condition item with a decision part, is used to choose condition item when generating rules. In primary and modified IGRs, two kinds of mutuality degree,simple and weighted mutuality are introduced respectively. In addition, the variable precision method is used to solve the conflict problem in modified IGRs. By experiments, the effects of two kinds heuristic knowledge and different weight values in synthetic mutuality on algorithms properties are shown,such as time consumption, calculation precision etc. The performances of IGRs with the primary and new conflict solution are compared by experiments. The conclusion is that the weighted mutuality degree is more sound and the choice of appropriate weight values in it are important to optimize the quality of rules set. The variable precision method for dealing with conflict when generating rules is more reasonable. Both two modifications to primary IGRs make the performance of IGRs enhanced and the quality of rules set better. Algorithm IGRs still need further improvement.

关键词： Criteria Dominance relation Algorithm IGRs Mutuality degree Conflict

来源：评论

学校读者我要写书评

暂无评论

Algorithm Studies of Rules Generation in CORS

Algorithm Studies of Rules Generation in CORS

引用

Firth IEEE International Conference on Cognitive Informatics(第五届认知信息国际会议)

作者： Chunxiao Fu Shuyang sun Chengmin Sun Dayou Liu College of Computer Science and Technology Jilin University Changchun 130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin Uni

ISBN: (纸本)1424404754

In Containing Order Rough Set Methodology (CORS), ordered attribute 'criterion' is introduced. Criterion is related semantically with decision attributes, which results in producing more rational and significant rules utilizing dominance relations. In this paper, some terminologies and properties on rules or rules set, such as robust, minimal, complete are discussed, the state of art in algorithms on rules generation are analyzed, and algorithms GRs and IGRs are proposed. We compare these two algorithms by experiments in time complexity, rules count and accuracy. We find that GRs can generate all minimal rules, but time complexity is high and IGRs' efficiency is better, but quality of rules are inferior to that of GRs. We also discuss the completeness of rules set and present a viewpoint that there are three hierarchies for completeness. In accuracy calculation, three formulas for accuracy calculation and two approaches for experimental test are given.

关键词： Criteria Dominance relation Algorithm GRs Algorithm IGRs Completeness

来源：评论

学校读者我要写书评

暂无评论

A hybrid discrete particle swarm algorithm for open-shop problems

引用

6th International Conference Simulated Evolution and Learning, SEAL 2006

作者： Qingyun, Yang Jigui, Sun Juyang, Zhang Chunjie, Wang College of Computer Science and Technology Jilin University Changchun 130012 China Key Laboratory for Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun 130012 China Open Laboratory for Intelligence Information Processing Fudan University Shanghai 200433 China Basic Sciences of ChangChun University of Technology ChangChun University of Technology Changchun 130012 China

ISBN: (纸本)3540473319

A hybrid discrete particle swarm algorithm is presented in this paper to solve open-shop problems. The operations are redefined in the discrete particle swarm algorithm. To improve the performance the simulated annealing algorithm is combined with discrete particle swarm. We use SA to enhance the results of local best positions instead of current positions. The experimental results show that our hybrid discrete particle swarm algorithm is effective and efficient to solve open-shop problems. © Springer-Verlag Berlin Heidelberg 2006.

关键词： Open systems

来源：评论

学校读者我要写书评

暂无评论

Visualization model of fuzzy region based on pleat set

引用

Xitong Fangzhen Xuebao / Journal of System Simulation 2006年第SUPPL.期18卷 340-342+346页

作者： Wu, De-Sheng Yu, Qiang-Yuan Institute of Information Spreading Engineering Changchun University of Technology Changchun 130012 China College of Computer Science and Technology Jilin University Changchun 130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun 130012 China

Fuzziness modeling for spatial data is currently an important problem in geographic information systems and spatial databases. In many geographical applications, spatial regions do not always have sharply defined boundaries but frequently their interiors and boundaries are fuzzy. A fuzzy spatial region model was proposed based on pleat set and the property of fuzzy regions was analyzed. The calculation method of relative membership of point in fuzzy region was given, and visualization model based on absolute membership value was proposed. The practical case shows that the model based on pleat set is valuable in the fields such as GIS, geography and spatial database.

关键词： Fuzzy region GIS Pleat set Visualization

来源：评论

学校读者我要写书评

暂无评论

Performance Evaluation of JXTA Based P2P Distributed Computing System

Performance Evaluation of JXTA Based P2P Distributed Computi...

引用

International Conference on Computing (CIC)

作者： Zhiming Dai Zhiyi Fang Xiao Han Fengyuan Xu Hongjun Yang School of Computer Science and Technology Key Laboratory Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University China School of Economics and Information Systems Jilin University China

In this paper, we introduce the condition of distributed computing at present firstly. On this foundation, according to the analysis of particular case of distributed computing network, we implement a distributed network environment of P2P whose bottom layer frame is based on JXTA. It improves issuing method of message based on pipeline decomposition mechanism of task based on usage ratio of processor and distributed mechanism of object based on serialization. We build an improved distributed computing network platform on the platform of Linux and Windows and simulate a distributed computing environment. The simulated experiment proves the feasibility and the validity of the distributed network computing platform that is constructed in this paper

关键词： Distributed computing Peer to peer computing Computer networks computational modeling Protocols Power generation economics Environmental economics Linux Large-scale systems Application software

来源：评论

学校读者我要写书评

暂无评论

Adaptive and iterative gene selection based on least squares support vector regression

引用

Journal of Information and computational Science 2006年第3期3卷 443-451页

作者： Jiang, Jingqing Wu, Chunguo Song, Chuyi Liang, Yanchun Key Laboratory of Symbol Computation and Knowledge Engineering College of Computer Science and Technology Jilin University Changchun 130012 China College of Mathematics and Computer Science Inner Mongolia University for Nationalities Tongliao 028043 China Key Laboratory of Information Science and Engineering of Railway Ministry The Key Laboratory of Advanced Information Science and Network Technology of Beijing Beijing Jiaotong University Beijing 100044 China

Cancer classification and identification are major areas in medical research. DNA microarrays could provide useful information for cancer classification at the gene expression level. The number of genes in a microarray is always several thousands while the number of training samples always several dozens. In such case most of the machine learning models suffer from the overfitting and it is necessary to select a handful of most informative genes. An adaptive and iterative gene selection algorithm based on least squares support vector machines is proposed in this paper. The algorithm adopts sequential forward selection search scheme. The number of selected genes can be determined adaptively. The total number of genes processed by the proposed algorithm is smaller than that processed by other algorithms using support vector machines. Results of numerical experiments show that the proposed algorithm trains fast and achieves comparable performance on two well-known benchmark problems.

关键词： Genes

来源：评论

学校读者我要写书评

暂无评论

WDM: A New Efficient Visualization Method of Classifying Web Documents Based on SOM

WDM: A New Efficient Visualization Method of Classifying Web...

引用

International Conference on computational Intelligence and Security

作者： Xi Bai Jigui Sun Hong Luo College of Computer Science and Technology Jilin University Changchun China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun China College of Information Science and Technology of Technology Beijing Institute of Technology Beijing China

ISBN: (纸本)1424406048

We propose a new visualization method WDM to classify documents by adding in the position-factors of words such as the title-factor and the first-sentence-factor based on a SOM neural network. We also discuss the selection of the function which is used to calculate the belong-to-probability in neurons' reflecting process. The experimental results indicate that WDM makes the boundaries of different documents greatly more clear, and thus it can produce more accurate and intuitive classification compared to the visualization methods which do not have position-factors considered

关键词： Wavelength division multiplexing Visualization Neurons Neural networks educational technology educational institutions Computer science Sun Laboratories Computer networks

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：