检索结果-内蒙古大学图书馆

A weighted semi-supervised dimensionality reduction method based locally linear graph

Journal of Information and Computational Science 2014年第13期11卷 4849-4826页

作者： Lu, Tianjun Liu, Shenglan Wang, Huibing Wang, Le School of Software Nanyang Institute of Technology Nanyang China School of Computer Science and Technology Faculty of Electronic Information and Electrical Engineering Dalian University of Technology Dalian China School of Innovation Experiment Dalian University of Technology Dalian China College of Information Engineering Ningbo Dahongying University Ningbo China

In recent years, dimensionality reduction has been widely used in image processing, pattern recognition, and other related fields, and achieved a good performance. Especially supervised dimensionality reduction algorithms have been successfully applied to preprocess of data in classification. However, in many practical applications, the size of labeled sample is generally too small for traditional supervised dimensionality reduction algorithms (such as Linear Discriminant Analysis, LDA) to overcome the Small Sample Size (SSS) problem;moreover, a small amount of labeled samples usually lead the projected data into an inappropriate low-dimensional subspace. In order to solve the two problems above, this paper proposes a Weighted Semi-supervised Maximum Margin Criterion (WSMMC) algorithm. Our method utilizes the Locally Linear Embedding Graph to characterize the spatial distribution of non-labeled samples and avoid the SSS problem by taking advantage of MMC. Experimental results show that the algorithm is effective. Copyright © 2014 Binary Information Press.

关键词： Discriminant analysis

来源：评论

学校读者我要写书评

暂无评论

A Dual-band Bandpass Filter Using A Transversal-interference Multimode Resonator

A Dual-band Bandpass Filter Using A Transversal-interference...

引用

Asia-Pacific Microwave Conference

作者： Dong Chen Wei Zhu Huizheng Bu Lei Zhu Chonghu Cheng School of Electronic Science and Engineering Nanjing University of Posts and Telecommunications Department of Electrical and Computer Engineering Faculty of Science and Technology University of Macau

ISBN: (纸本)9781479987689

A dual-band bandpass filter using a transversal-interference multimode resonator is proposed. The resonator is composed of a flatten ring resonator attached with two open-circuited stubs. By properly selecting the connecting position between the stubs and a ring, the two pairs of resonant modes can be grouped to make up the dual-passband response, and they can be further separated with virtue of two transmission zeros between them. By changing the length and the off-center position of these two stubs, both of dual passbands can be flexibly adjusted in terms of their central frequencies and bandwidths. In final, a dual-band filter with central frequencies at 3.5/5.4 GHz is designed and fabricated. The measured results are found in good agreement with the predicted ones.

关键词： Band-pass filters Dual band Centration Bandwidth Resonator Device Component virtues

来源：评论

学校读者我要写书评

暂无评论

Simulator of multi-service switching networks with multi-service sources 10

Simulator of multi-service switching networks with multi-ser...

引用

10th Advanced International Conference on Telecommunications, AICT 2014

作者： Gląbowski, Mariusz Krstić, Dragana Sobieraj, Maciej Department of Communication and Computer Networks Faculty of Electronics and Telecommunications Poznan University of Technology Poland Department of Telecommunications Faculty of Electronic Engineering University of Niš Serbia

ISBN: (纸本)9781612083605

In this paper, we present a simulator of multi-service switching networks with various resource management mechanisms. The network can be offered three types of traffic streams: Erlang, Engset and Pascal, generated by multi-service sources. Each multi-service source can generate calls of a number of traffic classes. In addition, the paper presents an analysis of the influence of different parameters upon the traffic characteristics of switching networks. Copyright © IARIA, 2014.

关键词： Switching networks

来源：评论

学校读者我要写书评

暂无评论

System Identification of a Prototype Small Scale ROV for Depth Control

System Identification of a Prototype Small Scale ROV for Dep...

引用

Asian Control Conference

作者： Mohd Shahrieel Mohd Aras Fadilah Abdul Azis Lim Wee Teck Shahrum Shah Abdullah Ahmad Fadzli Nizam Abdul Rahman Underwater Technology Research Group (UTeRG) Faculty of Electrical Engineering Universiti Teknikal Malaysia Melaka Department of Electronic System Engineering Malaysia-Japan International Institute of Technology Universiti Teknologi Malaysia KL Campus Faculty of Information and Computer Engineering Universiti Teknikal Malaysia Melaka

ISBN: (纸本)9781479978632

This paper present the design and development of a small scale underwater Remotely Operated Vehicle (ROV) and modelling the depth response of this ROV using System Identification Toolbox. The design of a small scale ROV has been done to minimize the hydrodynamic force and increase energy efficiency compared to the previous model that was developed by Underwater technology Research Group (UTeRG). The performance of the designed ROV will be tested in UTeRG laboratory (lab tank test). The output signal from the pressure sensor (MPX4250GP) and the Inertial Measurement Unit (IMU) sensor are interpreted via an NI-card which was used for the data transfer. The prototype ROV was compared with the previous version in terms of depth control performance. System identification toolbox in MATLAB was used to infer a model from open-loop experiments. Then the obtained model was used to design a controller for the ROV. The focus of the controller design will be to ensure that the ROV is stable and can maintain position at a certain depth in a real underwater environment. After all the experiment has been conducted, the ROV managed to operate in a certain depth underwater using the controller designed successfully.

关键词： Depth control Unmanned Underwater Remotely Operated Vehicle System Identification Toolbox

来源：评论

学校读者我要写书评

暂无评论

Measuring polarization with median filtering based on non-orthogonal angles 10

Measuring polarization with median filtering based on non-or...

引用

Joint International Conference of the 10th China-Japan International Workshop on Information technology and Control Applications and the 6th International Symposium on Computational Intelligence and Industrial Applications, ITCA and ISCIIA 2014

作者： Pang, Liyan Lv, Yange Zhu, Guohun School of Computer Science and Engineering Guilin University of Electronic Technology No.1 Jinji Road Guilin541004 China Faculty of Health Engineering and Sciences University of Southern Queensland ToowoombaQLD Australia

The measurement of degree of polarization strictly requires three orthogonal angles. The non-orthogonal positions can cause remarkable error based on Stokes operators. This paper proposes a median filtering pre-processing method to estimate the polarization degrees with non-orthogonal angle. Firstly, two group polarization positions are acquired from reflect light on key surface with a CCD camera, then the polarization degree images are analyzed and synthesized. By compared with the polarization degree from conventional algorithm, the accuracy of the calculation result improved 97%. This method can choose any three non-orthogonal angles, which is significantly improve to recognize the degree of polarization measurement in dynamic objects.

关键词： Polarization

来源：评论

学校读者我要写书评

暂无评论

Multiple mobile data offloading through disruption tolerant networks

引用

IEEE Transactions on Mobile Computing 2014年第7期13卷 1579-1596页

作者： Li, Yong Qian, Mengjiong Jin, Depeng Hui, Pan Wang, Zhaocheng Chen, Sheng Tsinghua National Laboratory for Information Science and Technology Department of Electronic Engineering Tsinghua University Beijing 100084 China Department of Computer Science and Engineering Hong Kong University of Science and Technology Hong Kong Telekom Innovation Laboratories Berlin Hong Kong Aalto University Helsinki Finland Electronics and Computer Science University of Southampton Southampton SO17 1BJ United Kingdom Faculty of Engineering King Abdulaziz University Jeddah 21589 Saudi Arabia

To cope with explosive traffic demands on current cellular networks of limited capacity, Disruption Tolerant Networking (DTN) is used to offload traffic from cellular networks to high capacity and free device-to-device networks. Current DTN-based mobile data offloading models are based on simple and unrealistic network assumptions which do not take into account the heterogeneity of mobile data and mobile users. We establish a mathematical framework to study the problem of multiple-type mobile data offloading under realistic assumptions, where (i) mobile data are heterogeneous in terms of size and lifetime;(ii) mobile users have different data subscribing interests;and (iii) the storages of offloading helpers are limited. We formulate the objective of achieving maximum mobile data offloading as a submodular function maximization problem with multiple linear constraints of limited storage, and propose three algorithms, suitable for the generic and more specific offloading scenarios, respectively, to solve this challenging optimization problem. We show that the designed algorithms effectively offload data to the DTN by using both the theoretical analysis and simulation investigations which employ both real human and vehicular mobility traces. © 2002-2012 IEEE.

关键词： Digital storage

来源：评论

学校读者我要写书评

暂无评论

A New Expert System Shell in Turkish Language for Training

A New Expert System Shell in Turkish Language for Training

引用

International Conference on challenges in IT, engineering and technology (ICCIET’2014, Phuket)

作者： Kemal Tutuncu Murat Koklu Electric-Electronic Engineering Department Technology Faculty of Selcuk University Computer Engineering Department Technology Faculty of Selcuk University

The aim of this study is to design an Expert System(ES) Shell in Turkish language for training license students about building ES. All the current ES shells and tools are required to know English and/or other foreign language. Since most of Turkish university present education in Turkish language and students are not good at any other foreign language, It is vital them to use education materials in Turkish language. The software developed for this study will help lecturer to teach ES, ES application and ES construction in a specific field. The developed ES shell has rule based knowledgebase, forward-chain inference mechanism and certainty factor for including fuzzy logic. The developed ES shell has user-friendly Graphical User Interface where all the menu items and tools are presented in Turkish language. In order to build a new ES in any area It facilitate easy entrance of domain specific variables, range and values that variables can have and rules that uses these variables. Apart from building new ESs in different areas the students will learn structural parts of ES(knowledge-base, working memory, inference mechanism, and etc.) by looking at the source code of the prepared ES shell.

关键词： ES Shell ES application Fuzzy Logic Knowledgebase Inference Mechanism

来源：评论

学校读者我要写书评

暂无评论

Qualitative Bankruptcy Prediction Rules Using Artificial Intelligence Techniques

Qualitative Bankruptcy Prediction Rules Using Artificial Int...

引用

International Conference on challenges in IT, engineering and technology (ICCIET’2014, Phuket)

作者： Murat Koklu Kemal Tutuncu Computer Engineering Department Technology Faculty of Selcuk University Electric-Electronic Engineering Department Technology Faculty of Selcuk University

Qualitative bankruptcy prediction rules represent experts' problem-solving knowledge to predict qualitative bankruptcy. The objective of this research is predicting qualitative bankruptcy using 4 different Artificial Intelligence(AI) techniques Qualitative Bankruptcy namely;Naive Bayes Classifier(NBC), Multilayer Perceptron(MLP), J48 and Classification via Regression(CR). Correctly Classified Instances were found as 96.5714 %, 94.8571 %, 95.4286 % and 96% for NBC, MLP, J48 and CR, respectively. These results have shown that NBC has the most successful prediction ratio among the four techniques regarding to classification. By using NBCs we can generate better rules with more qualitative factors and redundancy and overlapping of the rules can also be avoided.

关键词： Qualitative Bankruptcy Prediction Naive Bayes Classifier(NBC) Multi-Layer Perceptron(MLP) J48 and Classification via Regression(CR)

来源：评论

学校读者我要写书评

暂无评论

Determination of direction and speed of a uniformly moving obstacle from intersections with virtual window

Determination of direction and speed of a uniformly moving o...

引用

2014 IEEE Symposium on Industrial electronics and Applications, ISIEA 2014

作者： Mansor, Asri Hj. Rashid, Zarina Abdul Abdullah, Erma Rahayu Mohd Faizal Faculty of Electrical Engineering MARA University of Technology Shah Alam Malaysia Faculty of Electrical and Electronic Engineering University Tun Hussein Onn Malaysia Parit Raja Batu Pahat86400 Malaysia Dept. of AI Faculty of Computer Science and IT Malaya University Kuala Lumpur50603 Malaysia

ISBN: (纸本)9781479955909

A virtual window is used to determine the path and speed of a uniformly moving obstacle. Two intersections with the virtual window at different location are used to calculate the relative path and speed of the obstacle. Two simulations were performed using Excel 2010. The first simulation simulates a practical running of a uniformly moving obstacle, while the second has the obstacle moving at a very high speed. © 2014 IEEE.

关键词： Speed

来源：评论

学校读者我要写书评

暂无评论

Integrating Active Learning with Supervision for Crowdsourcing Generalization

Integrating Active Learning with Supervision for Crowdsourci...

引用

International Conference on Machine Learning and Applications (ICMLA)

作者： Zhenyu Shu Victor S. Sheng Yang Zhang Dianhong Wang Jing Zhang Heng Chen Institute of Geophysics & Geomatics South-Central University for Nationalities Wuhan China Department of Computer Science University of Central Arkansas Conway AR USA State Key laboratory of Networking an d Switching Technology Beijing University of Posts & Telecommunications Beijing China Faculty of Mechanical & Electronic Information China University of Geosciences Wuhan China School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China Shanghai Information Center for Life Sciences The Chinese Academy of Sciences Shanghai China

With various online crowdsourcing platforms, it is easy to collect multiple labels for the same examples from the crowd. Consensus integration algorithms can infer the estimated ground truths from the multiple label sets of these crowdsourcing datasets. However, it couldn't be avoided that these integrated (estimated) labels still contain noises. In order to further improve the performance of a model learned from data with these integrated labels, we propose an active learning framework to further improve the data quality, such that to improve the model quality, through acquiring limited true labels from experts (the oracle). We further investigate two active learning strategies in terms of two uncertainty measures (i.e., CLUE and MUE) within the active learning framework. From our experimental results on eight simulation crowdsourcing datasets and four real-world crowdsourcing datasets with three popular consensus integration algorithms, we draw several conclusions as follows. (i) Our active learning framework with the input from the oracle significantly improves the generalization ability of the model learned from crowdsourcing data. (ii) Our two active learning strategies outperform a random active learning strategy.

关键词： Crowdsourcing Uncertainty Data models Noise measurement Estimation Training Labeling

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：