检索结果-内蒙古大学图书馆

multi-objective evolutionary Neural Network to Predict Graduation Success at the United States Military Academy

Procedia Computer Science 2018年 140卷 196-205页

作者： Gene Lesinski Steven Corns United States Military Academy West Point NY 10996 USA Missouri University of Science and Technology Rolla MO 10996 USA

This paper presents an evolutionary neural network approach to classify student graduation status based upon selected academic, demographic, and other indicators. A pareto-based, multi-objective evolutionary algorithm utilizing the Strength Pareto evolutionary Algorithm (SPEA2) fitness evaluation scheme simultaneously evolves connection weights and identifies the neural network topology using network complexity and classification accuracy as objective functions. A combined vector-matrix representation scheme and differential evolution recombination operators are employed. The model is trained, tested, and validated using 5100 student samples with data compiled from admissions records and institutional research databases. The inputs to the evolutionary neural network model are used to classify students as: graduates, late graduates, or non-graduates. Results of the hybrid method show higher mean classification rates (88%) than the current methodology (80%) with a potential savings of $130M. Additionally, the proposed method is more efficient in that a less complex neural network topology is identified by the algorithm.

关键词： evolutionary algorithms, Neural network multi-objective evolutionary algorithms enrollment management student retention

来源：评论

学校读者我要写书评

暂无评论

Bi-MOCK: A multi-objective evolutionary Algorithm for Bi-clustering with Automatic Determination of the Number of Bi-clusters 24th

Bi-MOCK: A Multi-objective Evolutionary Algorithm for Bi-clu...

引用

24th International Conference on Neural Information Processing (ICONIP)

作者： Bousselmi, Meriem Bechikh, Slim Hung, Chih-Cheng Ben Said, Lamjed Univ Tunis Comp Sci Dept SMART Lab Tunis Tunisia Kennesaw State Univ Marietta GA USA Anyang Normal Univ Anyang Peoples R China

ISBN: (纸本)9783319700939;9783319700922

Bi-clustering is one of the main tasks in data mining with many possible applications in bioinformatics, pattern recognition, text mining, just to cite a few. It refers to simultaneously partitioning a data matrix based on both rows and columns. One of the main issues in bi-clustering is the difficulty to find the number of bi-clusters, which is usually pre-specified by the human user. During the last decade, a new algorithm, called MOCK, has appeared and shown its performance in data clustering where the number of clusters is determined automatically. Motivated by the interesting results of MOCK, we propose in this paper a new algorithm, called Bi-MOCK, which could be seen as an extension of MOCK for bi-clustering. Like MOCK, Bi-MOCK uses the concept of multi-objective optimization and is able to find automatically the number of bi-clusters thanks to a newly proposed variable string length encoding scheme. The performance of our proposed algorithm is assessed on a set of real gene expression datasets. The comparative experiments show the merits and the outperformance of Bi-MOCK with respect to some existing recent works.

关键词： Bi-clustering Number of bi-clusters multi-objective evolutionary algorithms Variable-size encoding

来源：评论

学校读者我要写书评

暂无评论

multi-objective trajectory optimization of Space Manoeuvre Vehicle using adaptive differential evolution and modified game theory

引用

ACTA ASTRONAUTICA 2017年第Jul.期136卷 273-280页

作者： Chai, Runqi Savvaris, Al Tsourdos, Antonios Chai, Senchun Cranfield Univ Sch Aerosp Transport & Mfg Cranfield MK43 0AL Beds England Beijing Inst Technol Sch Automat Beijing 100081 Peoples R China

Highly constrained trajectory optimization for Space Manoeuvre Vehicles (SMV) is a challenging problem. In practice, this problem becomes more difficult when multiple mission requirements are taken into account. Because of the nonlinearity in the dynamic model and even the objectives, it is usually hard for designers to generate a compromised trajectory without violating strict path and box constraints. In this paper, a new multi objective SMV optimal control model is formulated and parameterized using combined shooting-collocation technique. A modified game theory approach, coupled with an adaptive differential evolution algorithm, is designed in order to generate the pareto front of the multi-objective trajectory optimization problem. In addition, to improve the quality of obtained solutions, a control logic is embedded in the framework of the proposed approach. Several existing multi-objective evolutionary algorithms are studied and compared with the proposed method. Simulation results indicate that without driving the solution out of the feasible region, the proposed method can perform better in terms of convergence ability and convergence speed than its counterparts. Moreover, the quality of the pareto set generated using the proposed method is higher than other multi-objective evolutionary algorithms, which means the newly proposed algorithm is more attractive for solving multi-criteria SMV trajectory planning problem.

关键词： Trajectory optimization Space Manoeuvre Vehicles Modified game theory Adaptive differential evolution multi-objective evolutionary algorithms

来源：评论

学校读者我要写书评

暂无评论

multi-objective reverse engineering of variability-safe feature models based on code dependencies of system variants

引用

EMPIRICAL SOFTWARE ENGINEERING 2017年第4期22卷 1763-1794页

作者： Assuncao, Wesley K. G. Lopez-Herrejon, Roberto E. Linsbauer, Lukas Vergilio, Silvia R. Egyed, Alexander Fed Univ Parana UFPR DINF CP 19081 BR-81531980 Curitiba Parana Brazil Fed Univ Technol Parana UTFPR COTSI Cristo Rei St 19 BR-85902490 Toledo Brazil ETS Dept Software Engn & IT Notre Dame St Ouest 1100 Montreal PQ H3C 1K3 Canada Johannes Kepler Univ Linz JKU ISSE Altenbergerstr 69 A-4040 Linz Austria

Maintenance of many variants of a software system, developed to supply a wide range of customer-specific demands, is a complex endeavour. The consolidation of such variants into a Software Product Line is a way to effectively cope with this problem. A crucial step for this consolidation is to reverse engineer feature models that represent the desired combinations of features of all the available variants. Many approaches have been proposed for this reverse engineering task but they present two shortcomings. First, they use a single-objective perspective that does not allow software engineers to consider design trade-offs. Second, they do not exploit knowledge from implementation artifacts. To address these limitations, our work takes a multi-objective perspective and uses knowledge from source code dependencies to obtain feature models that not only represent the desired feature combinations but that also check that those combinations are indeed well-formed, i.e. variability safe. We performed an evaluation of our approach with twelve case studies using NSGA-II and SPEA2, and a single-objective algorithm. Our results indicate that the performance of the multi-objective algorithms is similar in most cases and that both clearly outperform the single-objective algorithm. Our work also unveils several avenues for further research.

关键词： Reverse engineering Feature models multi-objective evolutionary algorithms Empirical evaluation

来源：评论

学校读者我要写书评

暂无评论

Pareto adaptive penalty-based boundary intersection method for multi-objective optimization

引用

INFORMATION SCIENCES 2017年 414卷 158-174页

作者： Ming, Mengjun Wang, Rui Zha, Yabing Zhang, Tao Natl Univ Def Technol Coll Informat Syst & Management Changsha 410073 Hunan Peoples R China Foshan Univ Math & Big Data Foshan 528000 Peoples R China

Penalty-based boundary intersection (PBI) method is a frequently used scalarizing method in decomposition based multi-objective evolutionary algorithms (MOEAs). It works well when a proper penalty value is provided, however, the determination of a suitable penalty value depends on the problem itself, more precisely, the Pareto optimal front (PF) shape. As the penalty value increases, the PBI method becomes less effective in terms of convergence, but is more capable of handling various PF shapes. In this study, a simple yet effective method called Pareto adaptive PBI (PaP) is proposed by which a suitable penalty value can be adaptively identified, which therefore can maintain fast convergence speed, meanwhile, leading to a good approximation of the PF. The PaP strategy integrated into the state-of-the-art decomposition algorithm, MOEA/D, denoted as MOEA/D-PaP, is examined on a set of multi-objective benchmarks with different PF shapes. Experimental results show that the PaP strategy is more effective than the weighted sum, the weighted Tcheby-cheff and the PBI method with (representative) fixed penalty values in general. In addition, the MOEA/D-PaP is examined on a real-world problem multi-objective optimization of a hybrid renewable energy system whose PF is unknown. The outcome of the experiment further confirms its feasibility and superiority. (C) 2017 Elsevier Inc. All rights reserved.

关键词： multi-objective evolutionary algorithms Scalarizing method Pareto adaptive PEI method Hybrid renewable energy system

来源：评论

学校读者我要写书评

暂无评论

A hybrid project scheduling and material ordering problem: Modeling and solution algorithms

引用

APPLIED SOFT COMPUTING 2017年 58卷 700-713页

作者： Zoraghi, N. Shahsavar, A. Niaki, S. T. A. Islamic Azad Univ Fac Ind & Mech Engn Dept Ind Engn Qazvin Iran Sharif Univ Technol Dept Ind Engn Tehran Iran

A novel combination of a multimode project scheduling problem with material ordering, in which material procurements are exposed to the total quantity discount policy is investigated in this paper. The study aims at finding an optimal Pareto frontier for a triple objective model derived for the problem. While the first objective minimizes the makespan of the project, the second objective maximizes the robustness of the project schedule and finally the third objective minimizes the total costs pertaining to renewable and nonrenewable resources involved in a project. Four well-known multi-objective evolutionary algorithms including non-dominated sorting genetic algorithm II (NSGAII), strength Pareto evolutionary algorithm II (SPEAII), multi objective particle swarm optimization (MOPSO), and multi objective evolutionary algorithm based on decomposition (MOEAD) solve the developed triple-objective problem. The parameters of algorithms are tuned by the response surface methodology. The algorithms are carried out on a set of benchmarks and are compared based on five performance metrics evaluating their efficiencies in terms of closeness to the optimal frontier, diversity, and variance of results. Finally, a statistical assessment is conducted to analyze the results obtained by the algorithms. Results show that the NSGAII considerably outperforms others in 4 out of 5 metrics and the MOPSO performs better in terms of the remaining metric. (C) 2017 Elsevier B.V. All rights reserved.

关键词： Material ordering multi-mode project scheduling multi-objective evolutionary algorithms Pareto frontier Total quantity discount

来源：评论

学校读者我要写书评

暂无评论

Many-objective robust decision making for water allocation under climate change

引用

SCIENCE OF THE TOTAL ENVIRONMENT 2017年 607卷 294-303页

作者： Yan, Dan Ludwig, Fulco Huang, He Qing Werners, Saskia E. Wageningen Univ & Res Water Syst & Global Change POB 47 NL-6700 AA Wageningen Netherlands Chinese Acad Sci Inst Geog Sci & Nat Resources Res Key Lab Water Cycle & Related Land Surface Proc 11A Datun Rd Beijing 100101 Peoples R China

Water allocation is facing profound challenges due to climate change uncertainties. To identify adaptive water allocation strategies that are robust to climate change uncertainties, a model framework combining many-objective robust decision making and biophysical modeling is developed for large rivers. The framework was applied to the Pearl River basin (PRB), China where sufficient flow to the delta is required to reduce saltwater intrusion in the dry season. Before identifying and assessing robust water allocation plans for the future, the performance of ten state-of-the-art MOEAs (multi-objective evolutionary algorithms) is evaluated for the water allocation problem in the PRB. The Borg multi-objective evolutionary algorithm (Borg MOEA), which is a self-adaptive optimization algorithm, has the best performance during the historical periods. Therefore it is selected to generate new water allocation plans for the future (2079-2099). This study shows that robust decision making using carefully selected MOEAs can help limit saltwater intrusion in the Pearl River Delta. However, the framework could perform poorly due to larger than expected climate change impacts on water availability. Results also show that subjective design choices from the researchers and/or water managers could potentially affect the ability of the model framework, and cause the most robust water allocation plans to fail under future climate change. Developing robust allocation plans in a river basin suffering from increasing water shortage requires the researchers and water managers to well characterize future climate change of the study regions and vulnerabilities of their tools. (C) 2017 Elsevier B.V. All rights reserved.

关键词： multi-objective evolutionary algorithms Climate uncertainties Robust decision making Pearl River basin

来源：评论

学校读者我要写书评

暂无评论

multi-objective social spider optimisation algorithm

引用

JOURNAL OF ENGINEERING-JOE 2018年第16期2018卷 1521-1527页

作者： Wang, Yanjiao Sun, Xiaonan Wang, Chao Tao, Huanhuan Northeast Elect Power Univ Sch Elect Engn 169 Changchun Rd Jilin 132011 Jilin Peoples R China

Considering that the social spider algorithm is still unable to solve the multi-objective optimisation problem, this study presents a multi-objective social spider optimisation algorithm. Firstly, a new normalised fitness value formula is proposed based on the multi-objective optimisation purposes, which is able to trade off the non-dominated rankings and crowded distances and evaluate individual strengths and weaknesses effectively;secondly, the gravitational factor is used in order to balance the impact of individual fitness and distance to individual performance, which improves the vibration perception ability of the calculation method as well;once again, the renewal pattern of the female and male population is improved in order to balance the convergence rate and population diversity of the algorithm;lastly, the environmental selection strategy which is based on cosine distance is proposed for female and male population renewal. Testing on the ZDT test set, experimental results show that, compared with the six representative multi-objective evolutionary algorithms, the proposed algorithm in this study has better distribution and better convergence performance.

关键词： optimisation evolutionary computation fitness value formula non-dominated rankings gravitational factor ZDT test set multi-objective evolutionary algorithms multi-objective social spider optimisation algorithm

来源：评论

学校读者我要写书评

暂无评论

A NOVEL DYNAMIC CROWDING DISTANCE BASED DIVERSITY MAINTENANCE STRATEGY FOR MOEAS 16

A NOVEL DYNAMIC CROWDING DISTANCE BASED DIVERSITY MAINTENANC...

引用

International Conference on Machine Learning and Cybernetics (ICMLC)

作者： Yang, Ling Guan, Yuyang Sheng, Weiguo Zhejiang Univ Technol Sch Comp Sci & Technol Hangzhou 310023 Zhejiang Peoples R China Hangzhou Normal Univ Dept Comp Sci Hangzhou 310036 Zhejiang Peoples R China

ISBN: (纸本)9781538604083

Preserving population diversity is crucial for the performance of multi-objective evolutionary algorithms (MOEAs). In this paper, we propose a novel dynamic crowding distance based diversity preserving strategy for MOEAs. In the proposed strategy, the crowding distance is calculated based on the degree of deviation of each individual to its adjacent neighbors, thus appropriately adjusting the individual's density according to its position. Further, a multi-individual deletion mechanism is introduced to improve the efficiency of the strategy. For evaluation purpose, we incorporate the proposed strategy into NSGA-II and test it on ten functions. Our results show that the proposed strategy is able to improve the performance of NSGA-II and the resulting algorithm outperforms related methods to be compared.

关键词： multi-objective evolutionary algorithms Diversity preserve Dynamic crowding distance

来源：评论

学校读者我要写书评

暂无评论

A comparative study of many-objective evolutionary algorithms for the discovery of software architectures

引用

EMPIRICAL SOFTWARE ENGINEERING 2016年第6期21卷 2546-2600页

作者： Ramirez, Aurora Raul Romero, Jose Ventura, Sebastian Univ Cordoba Dept Comp Sci & Numer Anal Cordoba 14071 Spain

During the design of complex systems, software architects have to deal with a tangle of abstract artefacts, measures and ideas to discover the most fitting underlying architecture. A common way to structure such complex systems is in terms of their interacting software components, whose composition and connections need to be properly adjusted. Along with the expected functionality, non-functional requirements are key at this stage to guide the many design alternatives to be evaluated by software architects. The appearance of Search Based Software Engineering (SBSE) brings an approach that supports the software engineer along the design process. evolutionary algorithms can be applied to deal with the abstract and highly combinatorial optimisation problem of architecture discovery from a multiple objective perspective. The definition and resolution of many-objective optimisation problems is currently becoming an emerging challenge in SBSE, where the application of sophisticated techniques within the evolutionary computation field needs to be considered. In this paper, diverse non-functional requirements are selected to guide the evolutionary search, leading to the definition of several optimisation problems with up to 9 metrics concerning the architectural maintainability. An empirical study of the behaviour of 8 multi- and many-objective evolutionary algorithms is presented, where the quality and type of the returned solutions are analysed and discussed from the perspective of both the evolutionary performance and those aspects of interest to the expert. Results show how some many-objective evolutionary algorithms provide useful mechanisms to effectively explore design alternatives on highly dimensional objective spaces.

关键词： Software architecture discovery Search based software engineering Many-objective evolutionary algorithms multi-objective evolutionary algorithms

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

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

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：