检索结果-内蒙古大学图书馆

Structure learning of bayesian network with bee triple-population evolution strategies

International Journal of Advancements in computing Technology 2011年第10期3卷 140-148页

作者： Zhu, Yungang Liu, Dayou Jia, Haiyang Huang, Yuxiao 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

Recently evolutionary search has been investigated by lots of researchers as method for structure learning of Bayesian network. In order to avoid the problem of premature convergence and enhance the learning accuracy, in this paper we integrated evolutionary way of bees and proposed triple-population evolution strategies to learn Bayesian network structure. During evolution, the best Bayesian network individual is considered as queen bee, and there are three populations: one stores and evolves elitist individuals, another stores random individuals to guarantee the diversity of individuals, some individuals from that two populations form the third population in which every individual is considered as drone and mated with queen bee to generate offspring. The proposed method has both exploitative and explorative ability. Experimental results show that the proposed method can converge to more accurate Bayesian network than standard genetic algorithm based learning method.

关键词： Bayesian networks

来源：评论

学校读者我要写书评

暂无评论

A New Evolutionary Computation Based Approach for Learning Bayesian Network

引用

Procedia engineering 2011年 15卷 4026-4030页

作者： Yungang Zhu Dayou Liu Haiyang Jia 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

Bayesian network is a popular tool for uncertainty process in Artificial Intelligence. In recent years, more and more attention has been paid to learning of Bayesian network. In this paper, we proposed a novel learning algorithm for Bayesian network based on (μ, λ)-Evolution Strategy, we present the encoding scheme and fitness function, designed the evolutionary operators of recombination, mutation and selection. Theoretical analysis and experimental results all demonstrate that the proposed method can learn the Bayesian network from data effectively.

关键词： Bayesian network Evolution strategy Evolutionary computation

来源：评论

学校读者我要写书评

暂无评论

Damaged solar cell detection based on gray-intensity wave transformation

引用

Procedia engineering 2011年 15卷 3808-3813页

作者： Xuanjing Shen Wei Wei Jianwu Long Qingji Qian Colledge 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 130012China Colledge of Physics Jilin University Changchun 130025 China

An automatic system to detect cracks of solar cells on satellite solar panel through camera has been proposed in this study. And a novel binarization method based on gray intensity wave transformation is also introduced to decrease the impact of non-uniform illumination on cell image. It adaptively classifies the pixel depending on local peak and trough on the gray intensity surface. After the cell image binareized, a strategy called “adjacency searching” is adopted to remove the gate lines. Then the ellipse fitting based on least squares is conducted on the contours of crack segments to get their angles, which are used to connect these segments into the whole crack. The experiment on 5000 cells in one panel had shown that the precision of our system has reached 98.5%, and its false alarm rate is less than 9%, which could meet the application requirements.

关键词： Satellite Solar cell Binarization Crack detection

来源：评论

学校读者我要写书评

暂无评论

Optimize diagnosability and sensor placement in discrete event system

引用

Journal of Convergence Information Technology 2011年第7期6卷 219-224页

作者： Jian, Zhao Dantong, Ouyang Xiaoyu, Wang Liming, Zhang 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 Computer Science and Technology Changchun University Changchun China

Diagnosability indicates that whether the fault can be detected in finite time, which is an important property in model based diagnosis. As diagnosis depends on the sensor placement and the modeling, it is hard to make a choice whether to place more sensors for testing or to compute more diagnosis pathways in practical application. In this paper, a method is proposed to resolve this by defining key point. The key points take priority on testing among the existing sensors. In these key points, the observation can be optimized and be diagnosis tested efficiently. Experimental result indicates that this method achieves efficient faults distinction and identification, and reduces the cost of sensors and the computational complexity of diagnostic in the case of normal behavior.

关键词： Discrete event simulation

来源：评论

学校读者我要写书评

暂无评论

Eliminating repeated constraint checks in MaxRPC algorithms

引用

ICIC Express Letters, Part B: Applications 2011年第5期2卷 1189-1194页

作者： Guo, Jinsong Li, Zhanshan Zhang, Liang Geng, Xuena Key Laboratory of Symbolic Computation and Knowledge Engineering for Ministry of Education Jilin University No. 2699 Qianjin Street Changchun 130012 China College of Computer Science and Technology Jilin University No. 2699 Qianjin Street Changchun 130012 China

Filtering techniques are used in Constraint Satisfaction Problems to remove all the local inconsistencies during a processing step or prune the search tree efficiently during search. Local consistencies are used as preprocessing filtering techniques. MaxRPC which can remove more values than arc consistency is not used as widely as arc consistency for its relatively expensive time cost. We discover that, in the algorithms enforcing maxRPC, there are repeated constraint checks which can affect the efficiency of the algorithm negatively. In this paper, we propose a method making use of a data structure of AC4 and the relation "stronger" between local consistencies to eliminate the repeated constraint checks. Since the repeated constraint checks can be eliminated and the time cost of enforcing maxRPC can be reduced, the applicable occasions of maxRPC should be reconsidered. © 2011 ISSN 2185-2766.

关键词： Constraint satisfaction problems

来源：评论

学校读者我要写书评

暂无评论

A Heuristic Genetic Process Mining Algorithm

A Heuristic Genetic Process Mining Algorithm

引用

International Conference on Computational Intelligence and Security

作者： Jiafei Li Jihong OuYang Mingyong Feng Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun China Jilin University Changchun Jilin CN Industry Applications Department Cattsoft Technology Company Limited Beijing China

The current GPM algorithm needs many iterations to get good process models with high fitness which makes the GPM algorithm usually time-consuming and sometimes the result can not be accepted. To mine higher quality model in shorter time, a heuristic solution by adding log-replay based crossover operator and direct/indirect dependency relation based mutation operator is put forward. Experiment results on 25 benchmark logs show encouraging results.

关键词： Data mining Genetics Heuristic algorithms Computational modeling Process control Industries Measurement

来源：评论

学校读者我要写书评

暂无评论

A novel framework for gene selection

引用

International Journal of Advancements in computing Technology 2011年第3期3卷 184-191页

作者： Zhao, Wei Wang, Gang Wang, Hong-bin Chen, Hui-ling Dong, Hao Zhao, Zheng-dong Network Information Center Changchun University of Science and Technology Changchun 130022 China College of Computer Science and Technology Jilin University Changchun China Key Laboratory of Symbolic Computation Knowledge Engineering of Ministry of Education Jilin University Changchun 130012 China

Microarray data are highly redundant and noisy, and most genes are believed to be uninformative with respect to studied classes, as only a fraction of genes may present distinct profiles for different classes of samples. This paper proposed a novel hybrid framework (NHF) for the classification of high dimensional microarray data, which combined information gain(IG), F-score, genetic algorithm(GA), particle swarm optimization(PSO) and support vector machines(SVM). In order to identify a subset of informative genes embedded out of a large dataset which is contaminated with high dimensional noise, the proposed method is divided into three stages. In the first stage, IG is used to construct a ranking list of features, and only 10% features of the ranking list are provided for the second stage. In the second stage, PSO performs the feature selection task combining SVM. F-score is considered as a part of the objective function of PSO. The feature subsets are filtered according to the ranking list from the first stage, and then the results of it are supplied to the initialization of GA. Both the SVM parameter optimization and the feature selection are dynamically executed by PSO. In the third stage, GA initializes the individual of population from the results of the second stage, and an optimal result of feature selection is gained using GA integrating SVM. Both the SVM parameter optimization and the feature selection are dynamically performed by GA. The performance of the proposed method was compared with that of the PSO based, GA based, Ant colony optimization (ACO) based and simulated annealing (SA) based methods on five benchmark data sets, leukemia, colon, breast cancer, lung carcinoma and brain cancer. The numerical results and statistical analysis show that the proposed approach is capable of selecting a subset of predictive genes from a large noisy data set, and can capture the correlated structure in the data. In addition, NHF performs significantly better than th

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

A Loss Assessment Design Method Based on SOA for Insurance Informationization

A Loss Assessment Design Method Based on SOA for Insurance I...

引用

International Conference on Digital Manufacturing and Automation (ICDMA)

作者： Niya Li Jian Zhang Wei Lv Chengye Jin Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University China Jilin University Changchun Jilin CN College of Computer Science and Technology Jilin University China LeJin Electronics Company Limited Jilin China

The loss assessment is an important operation of claim process in insurance industry. On the growing tide of making the insurance information system the in-depth support to optimizing operation and serving insurant, a methodological framework for the loss assessment is given based on SOA technology, Under the framework, the operation process design, the client design, the service design and the database design are given. These design results have been validated by an actual application system.

关键词： Insurance Databases Companies Information systems Service oriented architecture Design methodology

来源：评论

学校读者我要写书评

暂无评论

An adaptive fuzzy ant colony optimization for feature selection

引用

Journal of Computational Information Systems 2011年第4期7卷 1206-1213页

作者： Liu, Yuanning Wang, Gang Chen, Huiling Zhao, Zhengdong Zhu, Xiaodong Liu, Zhen 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 Network Center Changchun University of Science and Technology Changchun 130022 China Graduate School of Engineering Nagasaki Institute of Applied Science Japan

Standard pattern classifiers perform on all data features. Whereas, some of the features are redundant or irrelevant, which reduce prediction accuracy, and increase running time of classifier. The purpose of this study is to search optimal feature subset, in order to increase the classification performance. The feature selection problem differs from traditional optimization problem on the problem size, namely the searching space or the problem dimension. Thus it is easy to converge at a local optimum by means of the ant colony optimization (ACO) based method with static parameter settings, but the adaptive parameter control strategy can balance between exploration and exploitation in the search space to intelligently improve solutions and avoid premature convergence. Only little research has been reported to adjust parameters of ACO adaptively. This paper proposes a novel adaptive fuzzy ant colony optimization (AFACO) based method for feature selection. We design two fuzzy controllers to adjust parameters of ACO, namely pheromone evaporation rate and the number of ants. The performance of the proposed method is compared with that of the Standard ACO based, Standard particle swarm optimization (PSO) based and genetic algorithm (GA) based methods on 10 benchmark data sets, taken from UCI machine learning and StatLog databases. The experimental results show that the proposed method outperforms the other methods in prediction accuracy, and is efficient and effective for feature selection problem. Copyright © 2011 Binary Information Press.

关键词： Ant colony optimization

来源：评论

学校读者我要写书评

暂无评论

Verilog combinational equivalence checking based on smt constraint solver

引用

ICIC Express Letters, Part B: Applications 2011年第5期2卷 1087-1092页

作者： Zeng, Hailin Ouyang, Dantong Zhang, Yuehua Zhang, Liming Zhang, Guiling School of Computer Science and Technology Jilin University No. 2699 Qanjin Street Changchun 130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University No. 2699 Qanjin Street Changchun 130012 China Software College Jilin University No. 2699 Qanjin Street Changchun 130012 China

Most previous works on combinational equivalence checking use BDDs and other Boolean level representations to formulate and solve the problem, and therefore, not utilizing the word-level information inherently presents in the design. To overcome this limitation, word level proof techniques like Satis ability Module Theory (SMT) were proposed. The SMT can efficiently handle the designs containing word-level arithmetic operators and bit-level Boolean logic. In this paper, we present a technique checking the equivalence of two combinational circuits based on SMT constraint solver. The experimental results indicate that our method can utilize word-level information to speed up the process of verification and result in a decrease in memory usage. © 2011 ISSN 2185-2766.

关键词： Formal verification

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：