检索结果-内蒙古大学图书馆

1st International Conference on Health Informatics, HEALTHINF 2008

作者： Mendez, David Gil Johnsson, Magnus Paya, Antonio Soriano Fernandez, Daniel Ruiz Computing Technology and Data Processing University of Alicante Spain Lund University Cognitive Science Dept. of Computer Science Lund University Lund Sweden

ISBN: (纸本)9789898111166

In this article we evaluate the work out of artificial neural networks as tools for helping and support in the medical diagnosis. In particular we compare the usability of one supervised and two unsupervised neural network architectures for medical diagnoses of lower urinary tract dysfunctions. The purpose is to develop a system that aid urologists in obtaining diagnoses, which will yield improved diagnostic accuracy and lower medical treatment costs. The clinical study has been carried out using the medical registers of patients with dysfunctions in the lower urinary tract. The current system is able to distinguish and classify dysfunctions as areflexia, hyperreflexia, obstruction of the lower urinary tract and patients free from dysfunction.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Generation of accurate integral surfaces in time-dependent vector fields

Generation of accurate integral surfaces in time-dependent v...

引用

作者： Garth, Christoph Krishnan, Hari Tricoche, Xavier Bobach, Tom Joy, Kenneth I. Institute of Data Analysis and Visualization University of California Davis United States Computer Science Dept. Purdue University Geometric Algorithms Group University of Kaiserslautern

We present a novel approach for the direct computation of integral surfaces in time-dependent vector fields. As opposed to previous work, which we analyze in detail, our approach is based on a separation of integral surface computation into two stages: surface approximation and generation of a graphical representation. This allows us to overcome several limitations of existing techniques. We first describe an algorithm for surface integration that approximates a series of time lines using iterative refinement and computes a skeleton of the integral surface. In a second step, we generate a well-conditioned triangulation. Our approach allows a highly accurate treatment of very large time-varying vector fields in an efficient, streaming fashion. We examine the properties of the presented methods on several example datasets and perform a numerical study of its correctness and accuracy. Finally, we investigate some visualization aspects of integral surfaces. © 2008 IEEE.

关键词： Numerical methods

来源：评论

学校读者我要写书评

暂无评论

Automatic feature selection for anomaly detection

Automatic feature selection for anomaly detection

引用

1st ACM Workshop on AISec, AISec'08, Co-located with the 15th ACM computer and Communications Security Conference, CCS'08

作者： Kloft, Marius Brefeld, Ulf Düssel, Patrick Gehl, Christian Laskov, Pavel Technical University of Berlin Dept. of Computer Science Berlin Germany Fraunhofer Institute FIRST Intelligent Data Analysis Berlin Germany University of Tübingen Wilhelm-Schickard-Institute Germany

ISBN: (纸本)9781605582917

A frequent problem in anomaly detection is to decide among different feature sets to be used. For example, various features are known in network intrusion detection based on packet headers, content byte streams or application level protocol parsing. A method for automatic feature selection in anomaly detection is proposed which determines optimal mixture coeffcients for various sets of features. The method generalizes the support vector data description (SVDD) and can be expressed as a semi-innite linear program that can be solved with standard techniques. The case of a single feature set can be handled as a particular case of the proposed method. The experimental evaluation of the new method on unsanitized HTTP data demonstrates that detectors using automatically selected features attain competitive performance, while sparing practitioners from a priori decisions on feature sets to be used. © 2009 IEEE.

关键词： Intrusion detection

来源：评论

学校读者我要写书评

暂无评论

Linkage learning, rule representation, and the χ-ary extended compact classifier system

引用

11th International Workshops on Learning Classifier Systems, WLCS 2007

作者： Llorà, Xavier Sastry, Kumara Lima, Cláudio F. Lobo, Fernando G. Goldberg, David E. National Center for Supercomputer Applications University of Illinois Urbana-Champaign IL 61801 United States Illinois Genetic Algorithms Laboratory University of Illinois Urbana-Champaign IL 61801 United States Dept. of Electronics and Computer Science Engineering University of Algarve Campus de Gambelas Faro 8000-117 Portugal Intel Corp.

ISBN: (纸本)3540881379

This paper reviews a competent Pittsburgh LCS that automatically mines important substructures of the underlying problems and takes problems that were intractable with first-generation Pittsburgh LCS and renders them tractable. Specifically, we propose a χ-ary extended compact classifier system (χeCCS) which uses (1) a competent genetic algorithm (GA) in the form of χ-ary extended compact genetic algorithm, and (2) a niching method in the form restricted tournament replacement, to evolve a set of maximally accurate and maximally general rules. Besides showing that linkage exist on the multiplexer problem, and that χeCCS scales exponentially with the number of address bits (building block size) and quadratically with the problem size, this paper also explores non-traditional rule encodings. Gene expression encodings, such as the Karva language, can also be used to build χeCCS probabilistic models. However, results show that the traditional ternary encoding 0,1,# presents a better scalability than the gene expression inspired ones for problems requiring binary conditions. © 2008 Springer Berlin Heidelberg.

关键词： Gene expression

来源：评论

学校读者我要写书评

暂无评论

A new stretch-twist-fold model for fast dynamo

引用

PAMM 2008年第1期7卷

作者： Renzo L. Ricca Francesca Maggioni Dept. Mathematics Statistics Computer Science & Applications U. Bergamo Via dei Caniana 2 24127 Bergamo Italy

Preliminary results on a new Stretch-Twist-Fold (STF) kinematic model for fast dynamo are presented. The evolution is prescribed by equations that govern the simultaneous stretching, writhing and coiling of a magnetic flux-tube by diffeomorphism of the initial circular configuration. Simple estimates based on minimized magnetic energy show that exponential growth of the magnetic field is indeed possible. (© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

关键词：

来源：评论

学校读者我要写书评

暂无评论

Biosequence classification using sequential pattern mining and optimization

Biosequence classification using sequential pattern mining a...

引用

6th International Special Topic Conference on ITAB, 2007

作者： Fotiadis, D.I. Exarchos, T.P. Tsipouras, M.G. Papaloukas, C. IEEE Unit of Medical Technology and Intelligent Information Systems Dept. of Computer Science University of Ioannina Ioannina Greece Dept. of Biological Applications and Technology University of Ioannina Ioannina Greece

ISBN: (纸本)9781424418688

In this paper we present a methodology for bio sequence classification, which employs sequential pattern mining and optimization algorithms. In the first stage, a sequential pattern mining algorithm is applied to a set of biological sequences and the sequential patterns are extracted. Then, the score of each pattern with respect to each sequence is calculated using a scoring function and the score of each class under consideration is estimated. The scores of the patterns and classes are updated, multiplied by a weight. In the second stage an optimization technique is employed to calculate the weight values to achieve the optimal classification accuracy. The methodology is applied in the protein class and fold prediction problem. Extensive evaluation is carried out, using a dataset obtained from the Protein data Bank. © 2008 IEEE.

关键词： Proteins

来源：评论

学校读者我要写书评

暂无评论

Visual vehicle egomotion estimation using the Fourier-Mellin Transform

Visual vehicle egomotion estimation using the Fourier-Mellin...

引用

2007 IEEE Intelligent Vehicles Symposium, IV 2007

作者： Goecke, Roland Asthana, Akshay Pettersson, Niklas Petersson, Lars Vision Science Technology and Applications NICTA Canberra Research Laboratory Canberra ACT Australia Dept. of Information Engineering RSISE Australian National University Canberra ACT Australia Computer Science and Information Technology Dept. Jaypee Institute of Information Technology Noida India

ISBN: (纸本)1424410681

This paper is concerned with the problem of estimating the motion of a single camera from a sequence of images, with an application scenario of vehicle egomotion estimation. Egomotion estimation has been an active area of research for many years and various solutions to the problem have been proposed. Many methods rely on optical flow or local image features to establish the spatial relationship between two images. A new method of egomotion estimation is presented which makes use of the Fourier-Mellin Transform for registering images in a video sequence, from which the rotation and translation of the camera motion can be estimated. The Fourier-Mellin Transform provides an accurate and efficient way of computing the camera motion parameters. It is a global method that takes the contributions from all pixels into account. The performance of the proposed approach is compared to two variants of optical flow methods and results are presented for a real-world video sequence taken from a moving vehicle. ©2007 IEEE.

关键词： Vehicles

来源：评论

学校读者我要写书评

暂无评论

The rewriting algorithm for disjunctive queries between peer ontologies

The rewriting algorithm for disjunctive queries between peer...

引用

3rd International Conference on Semantics, Knowledge, and Grid, SKG 2007

作者： Qin, Biao Wang, Shan Dept. of Computer Science Renmin University of China Beijing 100872 China Key Laboratory of Data Engineering and Knowledge Engineering Ministry of Education Beijing 100872 China

ISBN: (纸本)0769530079

We study the problem of answering queries given a set of mappings between peer ontologies. In addition to the schema mapping between peer ontologies, there are axioms to give constraints to classes and properties. We propose a set of rules to build graphs for the axioms. Because the axioms have different properties, the generated graphs are classified into four sets. In each peer, its RDF/OWL query languages can support regular expressions. If it wants to be transitive along semantic paths in peer knowledge management systems, we must rewrite conjunctive and disjunctive queries between peers. Because conjunctive queries are well-understood, we focus on a novel algorithm to rewrite disjunctive queries along semantic paths based on the graphs. For all atoms of a disjunctive query, we consider its union as a set and find the maximum rewritings over peers through a graphical way. Finally we do extensive simulation experiments. The simulation results show our algorithm can generate more rewritings than the naive rewriting algorithm at each distance. © 2007 IEEE.

关键词： Ontology

来源：评论

学校读者我要写书评

暂无评论

data Mining based Query Processing using Rough Sets and Genetic Algorithms

Data Mining based Query Processing using Rough Sets and Gene...

引用

IEEE Symposium on Computational Intelligence and data Mining (CIDM)

作者： K G Srinivasa M Jagadish K R Venugopal L M Patnaik Dept. of Computer Science M S Ramaiah Institute of Technology Bangalore Dept. of Computer Science Bangalore University Bangalore Microprocessor Applications Laboratory Indian Institute of Science Bangalore India

The optimization of queries is critical in database management systems and the complexity involved in finding optimal solutions has led to the development of heuristic approaches. Answering data mining query involves a random search over large databases. Due to the enormity of the data set involved, model simplification is necessary for quick answering of data mining queries. In this paper, we propose a hybrid model using rough sets and genetic algorithms for fast and efficient query answering. Rough sets are used to classify and summarize the datasets, whereas genetic algorithms are used for answering association related queries and feedback for adaptive classification. Here, we consider three types of queries, i.e., select, aggregate and classification based data mining queries. Summary tables that are built using rough sets and analytical model of attributes are used to speed up select queries. Mining associations, building concept hierarchies and reinforcement of reducts are achieved through genetic algorithms. The experiments are conducted on three real-life data sets, which include KDD 99 Cup data, Forest Cover-type data and Iris data. The performance of the proposed algorithm is analyzed for both execution time and classification accuracy and the results obtained are good

关键词： data mining Query processing Rough sets Genetic algorithms database systems Feedback Aggregates Analytical models Iris Performance analysis

来源：评论

学校读者我要写书评

暂无评论

Feature selection for neonatal resuscitation management using case-based reasoning

Modelling, Measurement and Control C

引用

Modelling, Measurement and Control C 2007年第3-4期68卷 67-85页

作者： Datta, S.K. Ghosh, I. Samant, R.K. Computer Centre Visva-Bharathi University Santiniketan-731 235 Birbhum India Dept. of Computer Science Ananda Chandra College Jalpaiguri-735 101 India Expert Systems Laboratory Dept. of Computer Science and Applications North Bengal University Darjeeling-734 013 India

Neonatal resuscitation management is one of the most important requirements in neonatal care. For the development of an automated decision support system in an expert (ES) framework for the efficient management of neonatal problems, rule-based knowledge systems are more common. But, however, rule-based approach is criticized for certain reasons. Case-based reasoning (CBR), a graceful alternative of rule-based approach is gaining swift momentum as a relatively new technology. For the development of a knowledge based system using CBR approach, selection of prime features having significant contributions and fixation of weights to features are important. This work applies a modified model for case-based learning suggested by Ghosh and Samanta [1]. This is a mathematical model for prime feature selection and fixation of their contributions for a problem domain. This is considered to be the first step for developing a case-based decision support system for neonatal resuscitation management. Previous case studies as definite cases were used to train the system. After a reasonable number of case inputs the significant features get saturated indicating the limit of retained cases in the case memory. The system derives the saturation values of each feature weight. It is also observed that the system rules out the non-significant features. These relative feature weights were used in a case-based classifier system using synapse matrix. Case studies have been presented to validate the system.

关键词： Feature Selection

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：