检索结果-内蒙古大学图书馆

international conference on P2P, parallel, Grid, Cloud and Internet Computing (3PGCIC)

作者： Marco Pota Angela Fuggi Massimo Esposito Giuseppe De Pietro Institute for High Performance Computing and Networking (ICAR) National Research Council of Italy (CNR) Napoli Italy Institute for High Performance Computing and Networking (ICAR) National Research Council of Italy (CNR) Naples Italy

ISBN: (纸本)9781467394741

Question Classification is one of the key tasks of Cognitive Systems based on the Question Answering paradigm. It aims at identifying the type of the possible answer for a question expressed in natural language. Machine learning techniques are typically employed for this task, and exploit a high number of features extracted from labelled questions of benchmark training sets in order to achieve good classification results. However, the high dimensionality of the feature space often limits the possibility of applying more efficient classification approaches, due to high training costs. In this work, more compact sets of lexical and syntactic features are proposed to distinguish question classes. In particular, the widely used unigrams are substituted with a smaller number of features, extracted by modifying typical Natural Language processing procedures for question analysis. the accuracy values gained on a benchmark dataset by using these different sets of features are compared among them and with the state-of-the-art, taking into account the required complexity at the same time. the new sets of extracted features show a good trade-off between accuracy and complexity.

关键词： Feature extraction Taxonomy Syntactics Aspirin Support vector machines Natural language processing Semantics

来源：评论

学校读者我要写书评

暂无评论

Accelerating text-based plagiarism detection using GPUs

Accelerating text-based plagiarism detection using GPUs

引用

international conference on Industrial and Information Systems, ICIIS

作者： MAC Jiffriya MAC Akmal Jahan Hasindu Gamaarachchi Roshan G. Ragel Post Graduate Institute of Science University of Peradeniya Peradeniya Sri Lanka Department of Computer Engineering University of Peradeniya Peradeniya Sri Lanka

Plagiarism is known as an unauthorized use of other's contents in writing and ideas in thinking without proper acknowledgment. there are several tools implemented for text-based plagiarism detection using various methods and techniques. However, these tools become inefficient while handling a large number of datasets due to the process of plagiarism detection which comprises of a lot of computational tasks and large memory requirement. therefore, when we deal with a large number of datasets, there should be a way to accelerate the process by applying acceleration techniques to optimize the plagiarism detection. In response to this, we have developed a parallel algorithm using Compute Unified Device Architecture (CUDA) and tested it on a Graphics processing Unit (GPU) platform. An equivalent algorithm is run on CPU platform as well. From the comparison of the results, CPU shows better performance when the number and the size of the documents are small. Meantime, GPU is an effective and efficient platform when handling a large number of documents and high in data size due to the increase in the amount of parallelism. It was found out that for our dataset, the performance of the algorithm on the GPU platform is approximately 6x faster than CPU. thus, introducing GPU based optimization algorithm to the plagiarism detection gives a real solution while handling a large number of data for inter-document plagiarism detection.

关键词： Acceleration Instruction sets Phased arrays Graphics processing units Kernel parallel processing Yttrium

来源：评论

学校读者我要写书评

暂无评论

Beyond Databases, architectures, and Structures: 10th international conference, BDAS 2014, Ustron, Poland, May 27-30, 2014. Proceedings 1

引用

丛书名： Communications in Computer and Information Science 424

2014年

作者： Stanislaw Kozielski, Dariusz Mrozek, Pawel Kasprowski, Bożena Małysiak-Mrozek, Daniel Kostrzewa (eds.)

ISBN: (数字)9783319069326

ISBN: (纸本)9783319069319;9783319069326

this book constitutes the refereed proceedings of the 10th IEEE international conference Beyond Databases, architectures, and Structures, BDAS 2014, held in Ustron, Poland, in May 2014. this book consists of 56 carefully revised selected papers that are assigned to 11 thematic groups: query languages, transactions and query optimization; data warehousing and big data; ontologies and semantic web; computational intelligence and data mining; collective intelligence, scheduling, and parallel processing; bioinformatics and biological data analysis; image analysis and multimedia mining; security of database systems; spatial data analysis; applications of database systems; Web and XML in database systems.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Design Space Exploration for a Co-designed Accelerator Supporting Homomorphic Encryption

Design Space Exploration for a Co-designed Accelerator Suppo...

引用

international conference on Control Systems and Computer Science (CSCS)

作者： Ghada Abozaid Ahmed El-Mahdy Parallel Computing Lab. Egypt-Japan University of Science and Technology (E-JUST) Alexandria Egypt (on-Leave) Faculty of Engineering Alexandria University Egypt

Fully Homomorphic Encryption is currently a sound theoretical approach for cloud security, it is currently not practically used due to the tremendous computation requirements of multiplying very large, million-bit, operands. In this paper, we explore the design space of software/hardware (SW/HW) co-designed accelerator relying on integrating fast software multiplication algorithms with a configurable hardware multiplier. the multiplier is based on a modified serial-parallel multiplier design, in which School-Book is a special case. the paper conducts an analytic performance study, exploring key design space parameters as well as comparing with other design approaches in the literature. Based on an actual FPGA implementation, we estimate a power consumption of 10, Watt, and area-time-power of 20.20 billion transistor-sec-Watt, potentially allowing for promising scalability.

关键词： Hardware Software Software algorithms Algorithm design and analysis parallel processing Space exploration Mathematical model

来源：评论

学校读者我要写书评

暂无评论

Machine Cognition and the Integration of Emotional Response in the Monitoring of Mental Disorders

Machine Cognition and the Integration of Emotional Response ...

引用

international conference on P2P, parallel, Grid, Cloud and Internet Computing (3PGCIC)

作者： Philip Moore Hai Van Pham Bin Hu Hong Liu Tarik Qaseem School of Information Science and Engineering Shandong Normal University Jinan China School of Information Science and Engineering Lanzhou University Lanzhou China Sch. of Inf. Technol. & Commun. Hanoi Univ. of Sci. & Technol. Hanoi Vietnam Warwick Medical School University of Warwick Coventry United Kingdom

ISBN: (纸本)9781467394741

Computer science relies heavily on computational modeling and while the value of such an approach is generally recognized the methodological account of computational explanation is not up-to-date. In this paper we explore machine cognition with the creation of effective cognitive modeling and consider the elemental components that combine to create an effective cognitive model. the creation of such a model will enable the processing of information in intelligent context-aware systems while integrating emotion (more accurately stated as emotive response). We present a brief review of related research addressing cognitive science and machine cognition in which we identify the concept of self. Modeling is introduced with an overview of conceptual models and semiotics followed by consideration of implementation using a proposed approach based on fuzzy sets. We introduce depression as a use-case to illustrate the proposed approach and a general discussion where future directions for research and open research questions are considered. the paper closes with concluding observations. We posit that creating an effective cognitive model offers the potential to integrate emotive response and thereby improve context-aware systems in a broad and diverse range of domains and systems along with improvements in the levels of computational intelligence.

关键词： Computational modeling Cognition Unified modeling language Cognitive science Context modeling Artificial intelligence Information processing

来源：评论

学校读者我要写书评

暂无评论

Concurrent Nomadic and Bundle Search: A Class of parallel algorithms for Local Optimization 1

引用

10th international conference on parallel processing and Applied Mathematics (PPAM)

作者： Voglis, Costas Papageorgiou, Dimitrios G. Lagaris, Isaac E. Nodalpoint Syst LTD Athens Greece

ISBN: (数字)9783642551956

ISBN: (纸本)9783642551956

We present a family of algorithms for local optimization that exploit the parallel architectures of contemporary computing systems to accomplish significant performance enhancements. this capability is important for demanding real time applications, as well as, for problems with time-consuming objective functions. the proposed concurrent schemes namely nomadic and bundle search are based upon well established techniques such as quasi-Newton updates and line searches. the parallelization strategy consists of (a) distributed computation of an approximation to the Hessian matrix and (b) parallel deployment of line searches on different directions (bundles) and from different starting points (nomads). Preliminary results showed that the new parallel algorithms can solve problems in less iterations than their serial rivals.

关键词： parallel local search Nomadic search Concurrent search Nested parallelism Line search Quasi-Newton

来源：评论

学校读者我要写书评

暂无评论

Elliptic Solver Performance Evaluation on Modern Hardware architectures

Elliptic Solver Performance Evaluation on Modern Hardware Ar...

引用

10th international conference on parallel processing and Applied Mathematics (PPAM)

作者： Ciznicki, Milosz Kopta, Piotr Kulczewski, Michal Kurowski, Krzysztof Gepner, Pawel Poznan Supercomp & Networking Ctr Noskowskiego 10 St PL-61704 Poznan Poland Intel Corp Swindon SN3 1RJ Wilts England

ISBN: (纸本)9783642552243

the recent advent of novel multi-and many-core architectures forces application programmers to deal with hardware-specific implementation details and to be familiar with software optimisation techniques to benefit from new high-performance computing machines. An extra care must be taken for communication-intensive algorithms, which may be a bottleneck for forthcoming era of exascale computing. this paper aims to present performance evaluation of preliminary adaptation techniques to hybrid MPI+OpenMP parallelisation schemes we provided into the EULAG code. Various techniques are discussed, and the results will lead us toward efficient algorithms and methods to scale communication-intensive elliptic solver with preconditioner, including GPU architectures to be provided later in the future.

关键词： Elliptic solver Preconditioning EULAG High performance computing Petascale computing Hybrid parallelisation Intel Xeon Phi

来源：评论

学校读者我要写书评

暂无评论

A Fast Ultrasonic Simulation Tool based on Massively parallel Implementations

A Fast Ultrasonic Simulation Tool based on Massively Paralle...

引用

10th international conference on Barkhausen and Micro-Magnetics (ICBM)

作者： Lambert, Jason Rougeron, Gilles Lacassagne, Lionel Chatillon, Sylvain CEA LIST F-91191 Gif Sur Yvette France Univ Paris 11 LRI F-91405 Orsay France

ISBN: (纸本)9780735412125

this paper presents a CIVA optimized ultrasonic inspection simulation tool, which takes benefit 01 the power of massively parallel architectures : graphical processing units (GPU) and multi-core general purpose processors (GPP). this tool is based on the classical approach used in CIVA : the interaction model is based on Kirchoff, and the ultrasonic field around the defect is computed by the pencil method. the model has been adapted and parallelized for both architectures. At this stage, the configurations addressed by the tool are : multi and mono-element probes. planar specimens made of simple isotropic materials, planar rectangular defects or side drilled holes of small diameter. Validations on the model accuracy and performances measurements are presented.

关键词： Ultrasonic Inspection Simulation Graphical processing Units Pencil Method Kirchhoff Model

来源：评论

学校读者我要写书评

暂无评论

Distributed XML processing over multicore servers

WEBIST 2014 - Proceedings of the 10th International Conferen...

引用

WEBIST 2014 - Proceedings of the 10th international conference on Web Information Systems and Technologies 2014年 1卷 200-207页

作者： Uratani, Yoshiyuki Koide, Hiroshi Cavendish, Dirceu Global Scientific Information and Computing Center Tokyo Institute of Technology O-okayama 2-12-1 Meguroku Tokyo 152-8550 Japan Faculty of Computer Science and Systems Engineering Kyushu Institute of Technology Kawazu 680-4 Iizuka Fukuoka 820-8502 Japan Network Design Research Center Kyushu Institute of Technology Kawazu 680-4 Iizuka Fukuoka 820-8502 Japan

ISBN: (纸本)9789897580239

Nowadays, multicore CPU become popular technology to enhance services quality in Web services. this paper characterizes parallel distributed XML processing which can off-load the amount of processing at their servers to networking nodes with varying number of CPU cores. Our implemented distributed XML processing system sends XML documents from a sender node to a server node through relay nodes, which process the documents before arriving at the server. When the relay nodes are connected in tandem, the XML documents are processed in a pipelining manner. When the relay nodes are connected in parallel, the XML documents are processed in a parallel fashion. For well-formedness and grammar validation tasks, the parallel processing reveals inherent advantages compared with pipeline processing regardless of document type, number of CPU cores and processing environment. Moreover, the number of CPU cores impacts efficiency of distributed XML processing via buffer access contention. Copyright © 2014 SCITEPRESS.

关键词： XML

来源：评论

学校读者我要写书评

暂无评论

Performance Characterization and Evaluation of HPC algorithms on Dissimilar Multicore architectures 16

Performance Characterization and Evaluation of HPC Algorithm...

引用

16th IEEE international conference on High Performance Computing and Communications HPCC 2014\11th IEEE international conference on Embedded Software and Systems ICESS 2014\6th international Symposium on Cyberspace Safety and Security CSS 2014

作者： Krishnan, S. P. T. Veeravalli, Bharadwaj Agcy Sci Technol & Res Inst Infocomm Res Singapore 138632 Singapore Natl Univ Singapore Dept Elect & Comp Engn Singapore 117583 Singapore

ISBN: (纸本)9781479961238

In this paper, we share our experiences in using two important yet different High Performance Computing (HPC) architectures for evaluating two HPC algorithms. the first architecture is an Intel x64 ISA based homogenous multicore with Uniform Memory Access (UMA) type shared-memory based Symmetric Multi-processing system. the second architecture is an IBM Power ISA based heterogenous multicore with Non-Uniform Memory Access (NUMA) based distributed-memory Asymmetric Multi-processing system. the two HPC algorithms are for predicting biological molecular structures, specifically the RNA secondary structures. the first algorithm that we created is a parallelized version of a popular serial RNA secondary structure prediction algorithm called PKNOTS. the second algorithm is a new parallel-by-design algorithm that we have developed called MARSs. Using real Ribo-Nucleic Acid (RNA) sequences, we conducted large-scale experiments involving hundreds of sequences using the above two algorithms. Based on thousands of data points that we collected as an outcome of our experiments, we report on the observed performance metrics for both the algorithms on the two architectures. through our experiments, we infer that architectures with specialized co-processors for number-crunching along with high-speed memory bus and dedicated bus controllers generally perform better than general-purpose multi-processor architectures. In addition, we observed that algorithms that are intrinsically parallelized by design are able to scale & perform better by taking advantage of the underlying parallel architecture. We further share best practices on handling scalability aspects with regards to workload size. We believe our results are applicable to other HPC applications on similar HPC architectures.

关键词： ARCHITECTURE High Performance Computing algorithms Multi-core processors Prediction algorithms parallel architectures Multi-processing Performance metrics Building construction Best Practices

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：