检索结果-内蒙古大学图书馆

Mining Hierarchical Decision Rules from Hybrid Data with Categorical and Continuous Valued Attributes

浙江海洋学院学报（自然科学版） 2010年第5期29卷 420-427页

作者： MIAO Duo-qian QIAN Jin LI Wen ZHANG Ze-hua Department of Computer＇Science and Technology Tongji University Shanghai 201804 China College of Computer Engineering Jiangsu Teachers University of Technology Changzhou 213015 China Key Laboratory of Embedded System and Service Computing Ministry of Education of China Tongji University Shanghai 201804 China

Decision rules mining is an important issue in machine learning and data ***,most proposed algorithms mine categorical data at single level,and these rules are not easily understandable and really useful for ***,a new approach to hierarchical decision rules mining is provided in this paper,in which similarity direction measure is introduced to deal with hybrid *** approach can mine hierarchical decision rules by adjusting similarity measure parameters and the level of concept hierarchy trees.

关键词： Similarity relation Attribute reduction Hierarchical decision rules Hybrid data

来源：评论

学校读者我要写书评

暂无评论

Decomposition based efficient algorithms for knowledge reduction

引用

ICIC Express Letters 2010年第3期4卷 1025-1032页

作者： Jiao, Na Miao, Duoqian Zhang, Hongyun Department of Computer Science and Technology Ministry of Education of China Tongji University Shanghai 201804 China Key Laboratory of Embedded System and Service Computing Ministry of Education of China Tongji University Shanghai 201804 China

Knowledge reduction, one of essential issues for data mining, has alwaysbeen a, hot topic due to the explosive growth of inform,a,tion. However, whenhandling large-scale data, m,any current knowledge reduction m,ethod,s based, onrough set theory are incapable. In this paper, efficient algorithms forknowledge reduction are put forward, based, on decomposition principle. Theidea, of decomposition is to break a, complex problem, down into a,m,a,ster-problem, and, several sub-problem,s that are simpler, m,orem,a,na,gea,ble and, m,ore solvable by using existing induction methods, thenjoining them, together in order to solve the original problem,. Compared, withsom,e traditional methods, the efficiency of the proposed algorithms can beillustrated by experiments with standard datasets from, UCI database. ICIC International © 2010.

关键词： Rough set theory

来源：评论

学校读者我要写书评

暂无评论

Improved asymptotic multicast throughput for random extended networks

引用

computer Communications 2010年第18期33卷 2195-2202页

作者： Cheng Wang Changjun Jiang Xiang-Yang Li Jiujun Cheng Department of Computer Science and Technology Tongji University Shanghai China State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications China Key Laboratory of Embedded System and Service Computing Ministry of Education Shanghai China Department of Computer Science Illinois Institute of Technology Chicago IL 60616 USA

We study the asymptotic throughput for random extended networks , where n ad hoc nodes are randomly deployed in a square region R ( n ) = 0 , n 2 . We directly consider the multicast throughput to unify the unicast and broadcast throughput, and design a new multicast scheme under the generalized physical model based on the so-called secondary highways system . Taking account of all possible cases of n s = ω (1) and 1 ⩽ n d ⩽ n − 1, we derive the achievable multicast throughput, where n s and n d denote the number of sessions and the number of destinations of each session. We prove that for some cases in terms of n s and n d , our scheme achieves better throughput than the existing schemes.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Coding Gabor Features for Multi-Modal Biometrics

Coding Gabor Features for Multi-Modal Biometrics

引用

Chinese Conference on Pattern Recognition (CCPR)

作者： Linlin Shen Zhen Ji Yuwen Li Li Bai Shenzhen Key Laboratory of Embedded System Design School of Computer Science & Software Engineering Shenzhen University China Shenzhen Institute of Information Technology Shenzhen China School of Computer Science University of Nottingham UK

Although biometrics technology has progressed substantially, its performance is still to be improved for real applications. This paper aims to improve the accuracy of personal identification, when only single sample is registered as template, by integrating multiple hand-based biometrics. To make fusion much easier, the same feature, so called fusion code, and decision level fusion strategy are used. Two fusion cases, face & palmprint and FKP & palmprint were taken as examples to verify the effectiveness. Experimental results show that much better performance than single modal biometrics has been achieved.

关键词： Face Biometrics Feature extraction Databases Testing Biomedical imaging Accuracy

来源：评论

学校读者我要写书评

暂无评论

Feature selection for semi-supervised multi-label learning with application to gene function analysis 10

Feature selection for semi-supervised multi-label learning w...

引用

Proceedings of the First ACM International Conference on Bioinformatics and Computational Biology

作者： Li, Guo-Zheng You, Mingyu Ge, Lei Yang, Jack Y. Yang, Mary Qu MOE Key Laboratory of Embedded System and Service Computing Tongji University Shanghai 201804 China Department of Control Science and Engineering Tongji University Shanghai 201804 China School of Computer Science and Engineering Shanghai University Shanghai 200072 China Purdue Univ. ECE West Lafayette IN 47907 United States Indiana Univ. Bioinformatics Center Indianapolis IN 46202 United States

ISBN: (纸本)9781450304382

This paper investigates gene function annotation of Yeast by using semi-supervised multi-label learning. Multi-label learning has been a hot topic in the bioinformatics field, but there are many samples unlabeled. Semi-supervised learning may be employed to utilize the unlabeled data. This paper proposes a novel semi-supervised multi-label learning algorithm COMN by combining Co-Training with ML-kNN to utilize the unlabeled yeast gene data to improve modeling accuracy of function annotation. Furthermore, an embedded feature selection algorithm PRECOMN is proposed to perform feature selection for COMN to remove the irrelevant and redundant features. Experimental results on one benchmark data set of Yeast show COMN and PRECOMN perform better than the original multi-label learning algorithm ML-kNN. Furthermore PRECOMN improves generalization performance of COMN. Copyright © 2010 ACM.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

L1/2 Sparsity Constrained Nonnegative Matrix Factorization for Hyperspectral Unmixing

L1/2 Sparsity Constrained Nonnegative Matrix Factorization f...

引用

Proceedings of the Digital Image Computing: Technqiues and Applications (DICTA)

作者： Yuntao Qian Sen Jia Jun Zhou Antonio Robles-Kelly College of Computer Science University of Zhejiang Hangzhou China Shenzhen City Key Laboratory of Embedded System Design College of Computer Science and Software Engineering Shenzhen University Shenzhen China Canberra Research Laboratory NICTA Canberra ACT Australia College of Engineering and Computer Science Australian National University Canberra ACT Australia

Hyperspectral unmixing is a crucial preprocessing step for material classification and recognition. In the last decade, nonnegative matrix factorization (NMF) and its extensions have been intensively studied to unmix hyperspectral imagery and recover the material end-members. As an important constraint, sparsity has been modeled making use of L 1 or L 2 regularizers. However, the full additivity constraint of material abundances is often overlooked, hence, limiting the practical efficacy of these methods. In this paper, we extend the NMF algorithm by incorporating the L 1/2 sparsity constraint. The L 1/2 -NMF provides more sparse and accurate results than the other regularizers by considering the end-member additivity constraint explicitly in the optimisation process. Experiments on the synthetic and real hyperspectral data validate the proposed algorithm.

关键词： Hyperspectral imaging Pixel Signal to noise ratio Equations Cost function Materials

来源：评论

学校读者我要写书评

暂无评论

Reduction Criteria for Probabilistic Models with Spatial Restrictions on States

Reduction Criteria for Probabilistic Models with Spatial Res...

引用

International Conference on Networks Security, Wireless Communications and Trusted Computing, NSWCTC

作者： Niu Jun Guosun Zeng Embedded System and Service Computing Key Laboratory Ministry of Education Shanghai China College of Information Engineering Zhejiang Business and Technology Institute Ningbo China Department of Computer Science and Technology University of Tongji Shanghai China

ISBN: (纸本)9781424465972;9780769540115

Partial order reduction techniques have been used to combat the state explosion problem in model checking procedures for concurrent systems with probabilistic behaviors. There are some results that give criteria on applying partial order reduction for verifying quantitative time properties and reward-based properties on actions. However, there are many situations that reward-based properties are expressed on states rather than on actions because actions are triggered in no time and the quantities can not be obtained easily. This paper presents reduction criteria for a probabilistic temporal logic that allows specification of restrictions on quantitative measures given by spatial resources function for the states of the considered system and provides the proof of the correctness.

关键词： Explosions computer networks Concurrent computing embedded computing Educational technology Probabilistic logic Communication system security computer security Information security Wireless communication

来源：评论

学校读者我要写书评

暂无评论

A Generalized Model with Spatial Resources Labels

A Generalized Model with Spatial Resources Labels

引用

International Conference on Biomedical Engineering and computer Science (ICBECS)

This paper proposes a generalized model by extending Markov chain with spatial resources labels, which can describe the functional and performance properties and some basic characteristics such as nondeterminacy and randomicity. The syntax and semantics of the new model are shown based on the existing temporal logics such as CSRL and pathCSRL. The model checking procedure is discussed and thus the practicability of the new model is intuitive. Based on the existing models, the proposed model includes spatial resources labels only by adding the spatial information to the existing labels of states, so the treatment of the state explosion problem is similar to that in the existing models.

关键词： Educational technology Paper technology Logic Labeling Educational institutions embedded system embedded computing computer science education systems engineering education computer science

来源：评论

学校读者我要写书评

暂无评论

Object rigidity and reflectivity identification based on motion analysis

Object rigidity and reflectivity identification based on mot...

引用

IEEE International Conference on Image Processing

作者： Di Zang Paul R. Schrater Katja Doerschner Department of Computer Science and Engineering University of Minnesota USA Key Laboratory of Embedded System and Service Computing Ministry of Education University of Tongji China National Research Center of Magnetic Resonance and Department of Psychology Bilkent University Turkey

Rigidity and reflectivity are important properties of objects, identifying these properties is a fundamental problem for many computer vision applications like motion and tracking. In this paper, we extend our previous work to propose a motion analysis based approach for detecting the object's rigidity and reflectivity. This approach consists of two steps. The first step aims to identify object rigidity based on motion estimation and optic flow matching. The second step is to classify specular rigid and diffuse rigid objects using structure from motion and Procrustes analysis. We show how rigid bodies can be detected without knowing any prior motion information by using a mutual information based matching method. In addition, we use a statistic way to set thresholds for rigidity classification. Presented results demonstrate that our approach can efficiently classify the rigidity and reflectivity of an object.

关键词： Optical imaging Adaptive optics Mutual information Shape computer vision Integrated optics Object recognition

来源：评论

学校读者我要写书评

暂无评论

Face recognition using gabor wavelet and self-adaptive intelligent single particle optimizer

Face recognition using gabor wavelet and self-adaptive intel...

引用

Chinese Conference on Pattern Recognition

作者： Zhou, Jiarui Ji, Zhen Huang, Weigan Tian, Tao College of Biomedical Engineering and Instrument Science Zhejiang University Hangzhou 310027 China College of Computer Science and Software Engineering Shenzhen University Shenzhen 518060 China Shenzhen Key Laboratory of Embedded System Design Shenzhen 518060 China

ISBN: (纸本)9781424472109

A Gabor wavelet based face recognition algorithm of PSOGabor is introduced in this paper. By using particular Gabor filters in feature extraction on important region of face image, PSOGabor can attain more representative feature information with less processing time. To overcome the drawback of overly dependence on the input parameters in intelligent single particle optimizer (ISPO), an improved algorithm of self-adaptive ISPO (AdpISPO) is proposed in this paper. With no critical parameter required, AdpISPO can achieve higher performance than original algorithms. Experimental results demonstrated that, by employing AdpISPO in Gabor filters' selection, PSOGabor can attain higher recognition rate with less computational time. ©2010 IEEE.

关键词： Face recognition

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：