检索结果-内蒙古大学图书馆

Implementing and Improving a method for non-invasive elicitation of probabilities for bayesian networks 07

Implementing and Improving a method for non-invasive elicita...

International Conference on computer Systems and Technologies and Workshop for PhD Students in Computing, CompSysTech '07

作者： De Jongh, Martinus Druzdzel, Marek Rothkrantz, Leon Faculty of Electrical Engineering Mathematics and Computer Science Man-Machine Interaction Group Delft University of Technology Delft Netherlands Decision Systems Laboratory School of Information Sciences and Intelligent Systems Program University of Pittsburgh Pittsburgh PA United States

ISBN: (纸本)9789549641509

Knowledge elicitation is difficult for expert systems that are based on probability theory. The elicitation of probabilities for a probabilistic model requires a lot of time and interaction between the knowledge engineer and the expert. Druzdzel and van der Gaag [5] have proposed a theoretical framework that would allow the use of multiple types of probabilistic information for elicitation of Bayesian networks. This framework has been used as a starting point for the implementation of a non-invasive elication method. © 2007 ACM.

关键词： Linear programming

来源：评论

学校读者我要写书评

暂无评论

Interpreting cabinet approval ratings by neural networks

Interpreting cabinet approval ratings by neural networks

引用

IASTED International Conference on Artificial Intelligence and Applications, AIA 2007

作者： Kamimura, Ryotaro Yoshida, Fumihiko Kitajima, Ryozo Information Science Laboratory Information Technology Center Tokai University 1117 Kitakaname Hiratsuka Kanagawa 259-1292 Japan Department of Media Studies Tokai University 1117 Kitakaname Hiratsuka Kanagawa 259-1292 Japan Graduate Program in Computer Sciences Tokai University 1117 Kitakaname Hiratsuka Kanagawa 259-1292 Japan

ISBN: (纸本)9780889866317

In this paper, we try to estimate Japan's cabinet approval ratings by using neural networks. In addition, we try to extract the important features in input patterns. This is the first attempt to use neural networks and to interpret the mechanism of inference for approval estimation in a comprehensive way. Experimental results show that neural networks have much better performance than that obtained by the standard regression analysis in terms of training and testing errors. The information loss analysis reveals that the first variable, that is, the previous ratings should play the most important role in inference. Though the experimental result here shown is a preliminary one, it certainly suggests a possibility of the automatic inference of cabinet approval ratings.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

information-Theoretic Variable Selection in Neural Networks

Information-Theoretic Variable Selection in Neural Networks

引用

IEEE Symposium on Foundations of Computational Intelligence, FOCI

作者： Ryotaro Kamimura Fumihiko Yoshida Yamashita Toshie Ryozo Kitajima Information Technology Center Information Science Laboratory Japan Department of Media Studies Tokai University Hiratsuka Kanagawa Japan Graduate Program in Computer Sciences Tokai University Hiratsuka Kanagawa Japan

In this paper, we propose a new type of information-theoretic approach to variable selection. Many approaches have been proposed in estimating the importance of input variables. The majority of these approaches have focused upon output errors. We here introduce an approach concerning internal representations. First, we delete an input unit with corresponding connection weights. Then, by examining some change in hidden unit activation with and without a input variable, we can extract an important variable. We apply this method to an artificial data in which the number of hidden units is redundantly increased so as to clearly show improved performance and the stability of our method. Then, we apply the method to the cabinet approval ratings in which better interpretation of input variables can be given

关键词： Input variables Neural networks Computational intelligence computer networks information theory Data mining Standards development Measurement units information science Laboratories

来源：评论

学校读者我要写书评

暂无评论

Delivering Context-aware Learning Guidance in the Mobile Learning Environment based on information Theory

Delivering Context-aware Learning Guidance in the Mobile Lea...

引用

International Conference on Advanced Learning Technologies (ICALT)

作者： Rita Kuo Min-Chang Wu Alex Chang Maiga Chang Jia-Sheng Heh Department of Electronic Engineer Chung Yuan Christian University Taiwan Department of Information and Computer Engineering Chung Yuan Christian University Taiwan Program Office of National Science and Technology Program for Learning in Taiwan Taiwan

Mobile Learning makes students get the advantages of both traditional learning and e-learning. How to give students learning sequence suggestions in the mobile learning environment is a big issue and is also the goal of this paper. This research uses knowledge map to store the characteristics of each learning object and designs a situated map to represent spatial knowledge in the mobile learning environment. By using these two knowledge structures, knowledge map and situated map, the system created by this research can generate various navigation sentences and ask students to observe the characteristics of learning objects. This research also takes information theory into consideration in order to decide which navigation sentence should deliver to the student first. The system calculates the entropy for each generated sentence and finds the most appropriate one to deliver to the student. At the end of this paper, an experiment system is implemented for the 5th year biology course, Plant Observation.

关键词： information theory Navigation Electronic learning Mobile computing Knowledge representation Humans Cognition Personal digital assistants US Department of Transportation Sequences

来源：评论

学校读者我要写书评

暂无评论

Quantitative and qualitative results of the national program for e-learning in Taiwan

引用

WSEAS Transactions on computers 2007年第3期6卷 546-551页

作者： Chang, Maiga Wang, Chin-Yeh Chen, Gwo-Dong National Science and Technology Program for e-Learning in Taiwan 300 Jhongda Rd. Chung-Li 32001 Taiwan Dept. of Computer Science and Information Engineering National Central Univ. 300 Jhongda Rd. Chung-Li 32001 Taiwan

This paper provides an overview of quantitative and qualitative outcome of the national program for e-learning in Taiwan (ELNP for short). The national program was planned and initiated in 2002 by the National science Council. Taiwan government wished to promote e-learning industry and make all people have same opportunity to learn knowledge via e-learning. A five-year national program for e-learning started from 2003. This paper shows both of quantitative and qualitative outcomes of national program in the past four years from four different ways: public welfare, academic researching output, e-learning industry, technological results.

关键词： E-learning

来源：评论

学校读者我要写书评

暂无评论

Sending the most appropriate feedbacks to students

引用

WSEAS Transactions on information science and Applications 2007年第3期4卷 584-591页

作者： Kuo, Rita Hsieh, Athena Chang, Maiga Heh, Jia-Sheng Department of Electronic Engineering Chung-Yuan Christian University No.200 Chung Pei Rd. Chung Li 32023 Taiwan Dep. of Information and Computer Engineering Chung-Yuan Christian University No.200 Chung Pei Rd. Chung Li 32023 Taiwan National Science and Technology Program for e-Learning in Taiwan No.300 Jungda-Rd. Chung-Li 32001 Taiwan

It is important for students getting suitable feedback information when either they are doing or completing a learning activity. This research tries to know what kind of feedback message is really useful to students. For this purpose, this paper develops a workable e-learning environment with feedback mechanism to gather the learning effects of students with different kinds of feedbacks. The experiment involves five fifth-grade classes and divides five classes into three different groups. The students in the control group have no feedback and students in the other two groups have different kinds of feedback messages. The analysis of experiment data contains two parts: the first one evaluates the feedback efficiencies for all groups;and, then all students will be separated into three clusters based on their academic achievement in order to do further analysis.

关键词： Feedback

来源：评论

学校读者我要写书评

暂无评论

Waters network's potential to transform environmental engineering education

Waters network's potential to transform environmental engine...

引用

114th Annual ASEE Conference and Exposition, 2007

作者： Eschenbach, Elizabeth Montgomery, Jami Johnson, James Brus, Chris Carlson, Patricia Giammar, Dan Grauer, Bette Hotaling, Liesl Oguntimein, Gbekeloluwa Safferman, Steven Wentling, Tim Humboldt State University United States Department of Environmental Resources Engineering Humboldt State University United States WATERS Network - CLEANER Project Office Network Howard University United States Department of Environmental Engineering College of Engineering Architecture and Computer Sciences Howard University United States American Academy of Environmental Engineers United States University of Iowa United States Program University of Iowa United States Rose-Hulman Institute of Technology United States Department of Rhetoric Rose-Hulman Institute of Technology United States National Research Council U. S. Air Force United States NASA's Classroom of the Future Wheeling WV United States Washington University United States Department of Energy Environmental and Chemical Engineering Washington University St. Louis United States Environmental Studies Program Center for Materials Innovation United States McPherson High School United States McPherson High School McPherson KS United States Stevens Institute of Technology United States Stevens Institute of Technology United States Morgan State University United States Michigan State University United States Biosystems and Agricultural Engineering Department Michigan State University United States University of Dayton United States Education Planning Committee National Center for Supercomputing Applications United States Department of Information Science Graduate School of Library and Information Science National Center for Supercomputing Applications University or Illinois United States Knowledge and Learning Systems Group NCSA United States

The WATERS Network (WATer and Environmental Research Systems Network) will be an integrated real-time distributed observing system which will enable academic and government scientists, engineers, educators, and practitioners to advance effective management of our nation's water resources by understanding human interactions with water and the natural and built environment. WATERS will provide easily accessible real time environmental data as well as analysis tools to engineers, scientists, educators, K- graduate students and policymakers so they can better understand how water quantity, quality and related components of the hydrologic cycle are impacted by natural and human influences. The WATERS strategic plan will be completed by July 2007 and requires constituent input. This paper describes the draft Conceptual Design and the Education Plan of WATERS Network. This paper will highlight the potential impact of WATERS on undergraduate and graduate environmental engineering education in order to elicit input from the environmental engineering education community on how WATERS Network could better meet the future needs of undergraduate and graduate students and educators. More information can be found at ***. © American Society for Engineering Education, 2007.

关键词： Water resources

来源：评论

学校读者我要写书评

暂无评论

Apriori data mining on rotationally invariant multiresolutional moments for pattern recognition

Apriori data mining on rotationally invariant multiresolutio...

引用

9th Joint Conference on information sciences, JCIS 2006

作者： Rodtook, A. Makhanov, S.S. Department of Computer Science Ramkhamhaeng University Thailand Information Technology and Computer Program Sirindhorn International Institute of Technology Thammasat University Thailand

ISBN: (纸本)9078677015

We propose a new feature selection procedure based on a combination of a pruning algorithm, Apriori mining techniques and fuzzy C-mean clustering. The feature selection algorithm is designed to mine on a multiresolution filter bank composed of rotationally invariant moments. The numerical experiments, with more than 10,000 images, demonstrate an accuracy increase of about 5% for a low noise, 15% for an average noise and 20% for a high-level noise.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

A treasure hunting learning model for students studying history and culture in the field with cellphone

A treasure hunting learning model for students studying hist...

引用

6th International Conference on Advanced Learning Technologies, ICALT 2006

作者： Chang, Alex Chang, Maiga Hsieh, Athena Department of Information and Computer Engineering Chung-Yuan Christian University National Science and Technology Program for E-learning Taiwan

ISBN: (纸本)0769526322

In the treasure hunting process, the hunters could only get few hints with simple words, therefore they must think over with the experience they learned before. The goal of this paper is to propose a treasure hunting learning model and implement a system based on the model. Within the treasure hunting learning model students will become a treasure hunter when they are learning in the field. The learning model makes learning more fun by asking students some questions according to their learning records and physical position. In order to make the system portable and treasure hunting liked, we implement the treasure hunting learning model with the most simple and common cellphone, that is the text-mode cellphone. To verify the learning model is working, a traditional history and culture course in college was taken as the experiment. © 2006 IEEE.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Customized benchmark generation using MDA

Customized benchmark generation using MDA

引用

5th Working IEEE/IFIP Conference on Software Architecture (WISCA 2005)

作者： Zhu, Liming Liu, Yan Gorton, Ian Bui, Ngoc Bao Empirical Software Engineering Program National ICT Australia Ltd. Australia School of Computer Science and Engineering University of New South Wales Australia Faculty of Information Technology University of Technology Sydney Australia

ISBN: (纸本)0769525482

This paper describes an approach for generating customized benchmark applications from a software architecture description using a Model Driven Architecture (MDA) approach. The benchmark generation and performance data capture tool implementation is based on widely used open source MDA frameworks. The business logic of the benchmark application is modeled in UML and generated by taking advantage of the existing generation "cartridges" so that the current component technology can be exploited in the benchmark. This greatly reduces the effort and expertise needed for benchmarking with complex component technology. We have also extended the MDA framework to model and generate a load testing suite and automatic performance measurement infrastructure. The approach complements current model-based performance prediction and analysis methods by generating the benchmark application from the same application architecture that the performance models are derived from. This provides the potential for tightly integrating runtime performance measurement with model-based prediction either for model validation or improving model prediction accuracy, We illustrate the approach using a case study based on EJB component technology.

关键词： Software architecture

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：