检索结果-内蒙古大学图书馆

Memetic pareto differential evolution for designing artificial neural networks in multiclassification problems using cross-entropy versus sensitivity

Memetic pareto differential evolution for designing artifici...

引用

4th International Conference on Hybrid Artificial Intelligence Systems, HAIS 2009

作者： Fernández, Juan Carlos Hervás, César Martínez, Francisco José Gutiérrez, Pedro Antonio Cruz, Manuel Department of Computer Science and Numerical Analysis University of Cordoba Campus de Rabanales Cordoba 14071 Spain Department of Management and Quantitative Methods ETEA Escritor Castilla Aguayo 4 Cordoba 14005 Spain

ISBN: (纸本)3642023185

This work proposes a Multiobjective Differential Evolution algorithm based on dominance Pareto concept for multiclassification problems using multilayer perceptron neural network models. The algorithm include a local search procedure and optimizes two conflicting objectives of multiclassifiers, a high correct classification rate and a high classification rate for each class, of which the latter is not usually optimized in classification. Once the Pareto front is built, we use two automatic selection methodologies of individuals: the best model with respect to accuracy and the best model with respect to sensitivity (extremes in the Pareto front). These strategies are applied to solve six classification benchmark problems obtained from the UCI repository. The models obtained show a high accuracy and a high classification rate for each class. © 2009 Springer Berlin Heidelberg.

关键词： Neural network models

来源：评论

学校读者我要写书评

暂无评论

Interpreting a Successful Testing Process:Risk and Actual Coverage

Interpreting a Successful Testing Process:Risk and Actual Co...

引用

Third International Symposium on Theoretical Aspects of Software Engineering TASE 2009(第三届软件工程理论国际研讨会)

作者： Mariělle Stoelinga Mark Timmer Formal Methods & Tools Group Department of Computer Science University of Twente The Netherlands

Testing is inherently incomplete;no test suite will ever be able to test all possible usage scenarios of a system. It is therefore vital to assess the implication of a system passing a test suite. This paper quantifies that implication by means of two distinct, but related, measures: the risk quantifies the confidence in a system after it passes a test suite, i.e., the number of faults still expected to be present (weighted by their severity);the actual coverage quantifies the extent to which faults have been shown absent, i.e., the fraction of possible faults that has been covered. We provide evaluation algorithms that calculate these metrics for a given test suite, as well as optimisation algorithms that yield the best test suite for a given optimisation criterion.

关键词： System testing Software testing Costs Software engineering computer science Solid modeling Mathematical model Error probability Random variables Risk management

来源：评论

学校读者我要写书评

暂无评论

Track-centric metrics for track fusion systems

Track-centric metrics for track fusion systems

引用

2009 12th International Conference on Information Fusion, FUSION 2009

作者： Canavan, Robert McCullough, Claire Farrell III, William J. Dept. of Computer Science and Engineering University of Tennessee at Chattanooga Chattanooga TN United States R. and D. Department Adaptive Methods Inc. Centreville VA United States

ISBN: (纸本)9780982443804

In order to reduce the labor-intensive nature of evaluating quality of fused tracks in a multi-sensor, multi-platform environment, the Metrics Assessment System (MAS) was developed to provide heavily automated, realtime metric calculation and display. The MAS presents metrics in tables and graphs within the tactical display and provides overlays on the tactical display for focused assessment. Originally, MAS included standard kinematic metrics, but more recent work has generated additional metrics based on track-to-track versions of Track Purity, Target Effectiveness, and Assignment Accuracy, metrics derived from a confusion matrix analysis of the fused and known truth tracks, and Receiver Operating Characteristic Curves. Data mining tools were used to assess the effectiveness of candidate metrics for inclusion in the MAS. This paper describes the metrics, the evaluations, and the framework for metric assessment which this work has provided. ©2009 ISIF.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

Lab-in-a-Box: Development of materials to support independent experimentation on concepts from circuits

Lab-in-a-Box: Development of materials to support independen...

引用

2009 ASEE Annual Conference and Exposition

作者： Meehan, Kathleen Hendricks, Robert Doolittle, Peter Clark, Richard Shek, Carl Virginia Tech United States Departments of Materials Science and Engineering and Electrical Computer Engineering Virginia Tech United States Department of Materials Science Department of Learning Science and Technology School of Education Virginia Tech United States Center for Excellence Undergraduate Teaching Virginia Western Community College United States

A project known as Lab-in-a-Box (LiaB) was developed in 2004 as one of the outcomes of a department-level reform within the Bradley department of Electrical and computer Engineering (ECE) at Virginia Tech, addressing a need that was identified through student and employer surveys for concrete examples of fundamental concepts in electrical engineering. LiaB is a set of 'hands-on' exercises in which students design, build, and test at home various d.c. and a.c. circuits using an inexpensive electronics kit, digital multimeter, and a software oscilloscope and, thus, has not require significant resources to implement. The inclusion of LiaB in our ECE curriculum has received overwhelmingly positive comments from the students as well as from faculty members who have used the kits for projects in upper division courses that have been traditionally lecture-based with no lab component and has been adopted by three community colleges. The aim of the first set of experiments that are under development, funded by a grant from the National science Foundation, is to reinforce abstract concepts on first-order and second-order RLC circuits introduced in the companion circuits lecture course. The students construct circuits with physical components rather than symbolic parts in PSpice and determine the time-varying voltage drops and currents in the circuit by direct measurement rather than by plugging values into their calculators. Experiments enable students to explore how the component tolerances, the initial state of the capacitor and/or inductor, and the frequency response of the circuit affects the output signal. An approach to integrate evaluation and assessment is being undertaken, where methods to measure the educational outcomes are considered concurrently with the development of the learning materials. A description of our pedagogical approach to the development of these learning materials and the integration of evaluation and assessment metrics will be described. © Americ

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Aligning computing education with engineering workforce computational needs: New curricular directions to improve computational thinking in engineering graduates

Aligning computing education with engineering workforce comp...

引用

39th Annual Frontiers in Education Conference: Imagining and Engineering Future CSET Education, FIE 2009

作者： Vergara, Claudia E. Urban-Lurain, Mark Dresen, Cindee Coxen, Tammy MacFarlane, Taryn Frazier, Kysha Briedis, Daina Buch, Neeraj Esfahanian, Abdol-Hossein Paquette, Louise Sticklen, Jon Laprad, Jeannine Wolff, Thomas F. Center for Research on College Science Teaching and Learning MSU United States Department of Instructional Technology Research and Development Division of Science and Mathematics Education College of Natural Science MSU United States Talent Development. CSW Australia Chemical Engineering and Materials Science MSU United States Civil and Environmental Engineering. MSU United States Computer Science and Engineering MSU United States Math and Computer Science Department LCC United States Engineering Education Research United States Undergraduate Studies College of Engineering MSU United States

ISBN: (纸本)9781424447152

In this global economy, the preparation of a globally competitive U.S. workforce with knowledge and understanding of critical computing concepts is essential. Our CPACE (Collaborative Process to Align Computing Education with Engineering Workforce Needs) vision is to revitalize undergraduate computing education within the engineering and technology fields. Our objective is to design and implement a process to engage stakeholders from multiple sectors and identify the computational tools and problem-solving skills and define how these skills-directly informed by industry needs-can be integrated across disciplinary curricula. By explicitly integrating computing concepts and disciplinary problem solving, engineering graduates will enter the workforce with improved and practice-ready computational thinking that will enhance their problemsolving and design skills. We present the analysis of the computational skills and the strategies that we are using to map the workforce problem-solving requirements onto the foundational computer science principles. We outline the framework that we are using to identify opportunities for curricular integration between computer science concepts and the disciplinary engineering curricula. By documenting, evaluating, and making the process explicit, this process can serve as a model for national efforts to strengthen undergraduate computing education in engineering. ©2009 IEEE.

关键词： Problem solving

来源：评论

学校读者我要写书评

暂无评论

Monitoring and Control of Large Systems with MonALISA

引用

COMMUNICATIONS OF THE ACM 2009年第9期52卷 49-70页

作者： Legrand, Iosif Voicu, Ramiro Cirstoiu, Catalin Grigoras, Costin Betev, Latchezar Costan, Alexandru Senior Research Engineer Caltech and The Technical Lead The Monalisa Project. Research Engineer at Caltech Working for Uslhcnet at CERN. Software Engineer The Finance Industry ALICE at CERN. Software Engineer in ALICE at CERN. The Offline Team of The ALICE Collaboration at CERN. Teaching Assistant The Computer Science Department of The University Politehnica of Bucharest.

The article focuses on the development of MonALISA, or Monitoring Agents using a Large Integrated Services Architecture, which is an application for controlling and monitoring large-scale distributed data-processing systems. The MonALISA framework, which was created by the California Institute of Technology's high energy physics group and has four layers of network look-up services, is explained. Research strategy focused on synergetics for computing, network infrastructure, applications, and storage facilities. The use of MonaALISA by the European Organization for Nuclear Research is mentioned.

关键词： DISTRIBUTED computing computer networks -- Monitoring INTELLIGENT agents (computer software) XML (Document markup language) SOFTWARE architecture SYNERGETICS PARTICLES (Nuclear physics) -- Software ROUTING (computer network management) PROGRAM transformation (computer programming) JAVA (computer program language)

来源：评论

学校读者我要写书评

暂无评论

Multilogistic Regression using Initial and Radial Basis Function covariates

Multilogistic Regression using Initial and Radial Basis Func...

引用

2009 International Joint Conference on Neural Networks, IJCNN 2009

作者： Gutiérrez, Pedro Antonio Hervás-Martinez, César Martínez-Estudillo, Francisco J. Fernández, Juan Carlos Department of Computer Science and Numerical Analysis University of Córdoba Campus de Rabanales Albert Einstein Building 14071 - Córdoba Spain Department of Management and Quantitative Methods ETEA Escritor Castilla Aguayo 4 14004 - Córdoba Spain

ISBN: (纸本)9781424435531

This paper proposes a hybrid multilogistic model, named MultiLogistic Regression using Initial and Radial Basis Function covariates (MLRIRBF). The process for obtaining the coefficients is carried out in several steps. First, an Evolutionary Programming (EP) algorithm is applied, aimed to produce a RBF Neural Network (RBFNN) with a reduced number of RBF transformations and the simplest structure possible. Then, the input space is transformed by adding the nonlinear transformations of the input variables given by the RBFs of the best individual in the last generation. Finally, a maximum likelihood optimization method determines the coefficients associated with a multilogistic regression model built on this transformed input space. In this final step, two different multilogistic regression algorithms are applied, one that considers all initial and RBF covariates (MLRIRBF) and another one that incrementally constructs the model and applies cross-validation, resulting in an automatic covariate selection (MLRIRBF*). The methodology proposed is tested using six benchmark classification problems from well-known machine learning problems. The results are compared with the corresponding multilogistic regression methodologies applied over the initial input space, to the RBFNNs obtained by the EP algorithm (RBFEP) and to other competitive machine learning techniques. The MLRIRBF* models are found to be better than the corresponding multilogistic regression methodologies and the RBFEP method for almost all datasets, and obtain the highest mean accuracy rank when compared to the rest of methods in all datasets. ©2009 IEEE.

关键词： Radial basis function networks

来源：评论

学校读者我要写书评

暂无评论

A synthesis course in hardware architecture, compilers, and software engineering

SIGCSE Bulletin Inroads

引用

SIGCSE Bulletin Inroads 2009年第1期41卷 443-447页

作者： Schocken, Shimon Nisan, Noam Armoni, Michal IDC Herzliya Efi Arazi School of Computer Science P.O.B. 167 Herzliya 46150 Israel School of Computer Science and Engineering Hebrew University of Jerusalem Jerusalem 91904 Israel Department of Science Teaching Weizmann Institute of Science Rehovot 76100 Israel

We describe a synthesis course that provides a hands-on treatment of many hardware and software topics learned in computer science (CS) programs. Using a modular series of twelve projects, we walk the students through the gradual construction of a simple hardware platform and a modern software hierarchy, yielding a basic yet powerful computer system. In the process of building the computer, the students gain a first-hand understanding of how hardware and software systems are designed and how they work together, as one enterprise. The course web site contains all the materials necessary to run this course in open source, and students and instructors are welcome to use and extend them freely. The course projects are modular and self-contained, and any subset of them can be implemented in any order and in any programming language. Therefore, they comprise a flexible library of exercises that can be used in many applied CS courses. This paper gives a description of the approach and the course, juxtaposed against general educational principles underlying meaningful learning.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Design of Artificial Neural Networks Using a Memetic Pareto Evolutionary Algorithm Using as Objectives Entropy versus Variation Coefficient

Design of Artificial Neural Networks Using a Memetic Pareto ...

引用

International Conference on Intelligent Systems Design and Applications (ISDA)

作者： J.C. Fernández C. Hervás F.J. Martínez M. Cruz Department of Computer Science University of Córdoba Cordoba Spain Department of Management and Quantitative Methods ETEA Cordoba Spain

This paper proposes a multi-classification pattern algorithm using multilayer perceptron neural network models which try to boost two conflicting main objectives of a classifier, a high correct classification rate and a high classification rate for each class. To solve this machine learning problem, we consider a memetic Pareto evolutionary approach based on the NSGA2 algorithm (MPENSGA2), where we defined two objectives for determining the goodness of a classifier: the cross-entropy error function and the variation coefficient of its sensitivities, because both measures are continuous functions, making the convergence more robust. Once the Pareto front is built, we use an automatic selection methodology of individuals: the best model in accuracy (upper extreme in the Pareto front). This methodology is applied to solve six benchmark classification problems, obtaining promising results and achieving a high classification rate in the generalization set with an acceptable level of accuracy for each class.

关键词： Algorithm design and analysis Artificial neural networks Evolutionary computation Entropy Machine learning algorithms Multilayer perceptrons Neural networks Multi-layer neural network Machine learning Convergence

来源：评论

学校读者我要写书评

暂无评论

Empirical usage metadata in learning objects

Empirical usage metadata in learning objects

引用

39th Annual Frontiers in Education Conference: Imagining and Engineering Future CSET Education, FIE 2009

作者： Nugent, Gwen Kupzyk, Kevin Riley, S.A. Miller, L.D. Hostetler, Jesse Soh, Leen-Kiat Samal, Ashok Nebraska Center for Research on Children Youth Families and Schools University of Nebraska-Lincoln United States Department of Educational Psychology Quantitative and Qualitative Psychometric Methods University of Nebraska-Lincoln United States Department of Computer Science and Engineering University of Nebraska-Lincoln United States

ISBN: (纸本)9781424447152

The iLOG Project (Intelligent Learning Object Guide) is designed to augment multimedia learning objects with information about (1) how a learning object has been used, (2) how it has impacted instruction and learning, and (3) how it should be used. The goal of the project is to generate metadata tags from data collected while students interact with learning objects;these metadata tags can then be used to help teachers identify learning objects that match the educational and experiential backgrounds of their students. The project involves the development of an agent-based intelligent system for tracking student interaction with learning objects, in tandem with an extensive learning research agenda. This paper provides an overview of this NSF-funded project, focusing on the instructional approach and research on varying levels of active learning and feedback. Using a randomized design and a hierarchical linear modeling framework, research showed that the active learning conditions resulted in significantly higher student learning. The elaborative feedback results approached (p = .056), but did not reach, the established significance criteria of alpha = .05. Both active learning conditions and one of the elaborative feedback conditions resulted in significantly higher content assessment scores compared to a control group. ©2009 IEEE.

关键词： Metadata

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：