检索结果-内蒙古大学图书馆

Proceedings - International Conference on Pattern Recognition 2002年第1期16卷 480-482页

作者： Xiaoyan, Zhu Xiaoxin, Yin IEEE State Key Lab. of Intelligent Tech. and Systems Dept. of Computer Science and Technology Tsinghua University Beijing 100084 China

A robust approach is proposed for document skew detection. We use Fourier analysis and SVM to classify textual areas from non-textual areas of documents. We also propose a robust method to determine the skew angle from textual areas. Our approach achieves good performance on documents with large area of non-textual contents. © 2002 IEEE.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Correlation-Based Web Document Clustering for Adaptive Web Interface Design

引用

Knowledge and Information systems 2002年第2期4卷 151-167页

作者： Su, Zhong Yang, Qiang Zhang, Hongjiang Xu, Xiaowei Hu, Yu-Hen Ma, Shaoping State Key Lab of Intelligent Tech. and Systems Tsinghua University Beijing China School of Computing Science Simon Fraser University BurnabyBC Canada Microsoft Research China Beijing China Siemens AG Information and Communications Corporate Technology Munich Germany Department of Electrical and Computer Engineering University of Wisconsin–Madison MadisonWI United States

A great challenge for web site designers is how to ensure users' easy access to important web pages efficiently. In this paper we present a clustering-based approach to address this problem. Our approach to this challenge is to perform efficient and effective correlation analysis based on web logs and construct clusters of web pages to reflect the co-visit behavior of web site users. We present a novel approach for adapting previous clustering algorithms that are designed for databases in the problem domain of web page clustering, and show that our new methods can generate high-quality clusters for very large web logs when previous methods fail. Based on the high-quality clustering results, we then apply the data-mined clustering knowledge to the problem of adapting web interfaces to improve users' performance. We develop an automatic method for web interface adaptation: by introducing index pages that minimize overall user browsing costs. The index pages are aimed at providing short cuts for users to ensure that users get to their objective web pages fast, and we solve a previously open problem of how to determine an optimal number of index pages. We empirically show that our approach performs better than many of the previous algorithms based on experiments on several realistic web log files. © 2002, Springer-Verlag London Limited.

关键词： Websites

来源：评论

学校读者我要写书评

暂无评论

A Fast Hybrid Algorithm of Global Optimization for Feedforward Neural Networks

引用

中国电子杂志（英文版） 2001年第2期10卷 214-218页

作者： JIANG Minghu ZHANG Bo ZHU Xiaoyan JINAG Mingyan Lab of Computational Linguistics Dept. of Chinese Language Tsinghua University Beijing 100084 China State Key Lab of Intelligent Tech. & Systems Dept. of Computer Tsinghua University Beijing 100084 China Dept. of Electronic Engineering Shandong University Jinan 250100 China

This paper presents the hybrid algorithm of global optimization of dynamic learning rate for multilayer feedforward neural networks (MLFNN).The effect of inexact line search on conjugacy was studied, based on which a generalized conjugate gradient method was proposed, showing global convergence for error backpagation of MLFNN. It overcomes the drawback of conventional BP and Polak-Ribieve conjugate gradient algorithms that maybe plunge into local minima. The hybrid algorithm's recognition rate is higher than that of Polak-Ribieve algorithm and convergence BP for test data, its training time is less than that of Fletcher-Reeves algorithm and far less than that of convergence BP, and it has a less complicated and stronger robustness to real speech data.

关键词： The conjugate gradient algorithm Global convergence Backpropagation.

来源：评论

学校读者我要写书评

暂无评论

Segmentation of Mandarin Braille Word and Braille Translation Based on Multi-knowledge

Segmentation of Mandarin Braille Word and Braille Translatio...

引用

2000年第16届世界计算机大会

作者： Jiang Minghu Zhu Xiaoyan Xia Ying Tan Gang Yuan Baozong Tang Xiaofang Institute of Information Sceince Northern Jiaotong University State Key Lab of Intelligent Tech.& Systems Department of ComputerTsinghua University

This paper is about the segmentation of Braille words and the transformation from Mandarin Braille to Chinese *** word segmentation consists of rules base,the signs base of segmentation and knowledge base for disambiguation and mistakes,by using adjacency constraints and bidirectional maximum matching with a dictionary,our system's segmentation precision is better than 99%for the common *** incorporating a pinyin knowledge dictionary into the system,we perfectly solved the problem of ambiguity in the translation from Braille to pinyin and developed a statistical language model based on the transformation of pinyin into *** using a multi-knowledge base to carry out the disambiguation process for each pinyin sentence,we built a multi-level graph and used Viterbi search to find the sequence of Chinese characters with maximum likelihood,and used an N-Best algorithm to get the N most likely character *** experimental results show that the system's overall precision for translation from Braille codes to Chinese characters is 94.38%.

关键词： Braille translation Virterbi algorithm smoothing method multi-knowledge

来源：评论

学校读者我要写书评

暂无评论

Analysis of the Effects of Quantization in High-Order Function Neural Network

Analysis of the Effects of Quantization in High-Order Functi...

引用

2000年第16届世界计算机大会

作者： Jiang Minghu Zhu Xiaoyan Lin Ying Yuan Baozong Tang Xiaofang Lin Biqin Ruan Qiuqi Jiang Mingyan Institute of Information Sceince Northern Jiaotong University State Key Lab of Intelligent Tech.& Systems Dept.of ComputerTsinghua University Dept.of Electronic Shandong University

A statistical quantization model is used to analyze of the effects of quantization in digital implementment of high-order function neural *** the theory we analysis the performance degradation and fault tolerance ability of the neural network caused by number of quantization bits and changing of *** try to predict the error in HOFNN given the properties of the network and the number of bits of quantization. Experiment results show the error rate is inverse proportional to quantized bits M for *** recognition performance of BP network and HRFNN are almost same for different quantization *** network's performance degradation gets worse when the number of bits is lower than 4-bit *** network's performance degradation gets worse when the number of bits is lower than 4-bit quantization.

关键词： Effects of Quantization high-order function neural networks statistical modelⅠ.Introduction

来源：评论

学校读者我要写书评

暂无评论

A Fast Hybrid Algorithm of Global Optimization for Feedforward Neural Networks

A Fast Hybrid Algorithm of Global Optimization for Feedforwa...

引用

2000年第16届世界计算机大会

作者： Jiang Minghu Zhu Xiaoyan Yuan Baozong Tang Xiaofang Lin Biqin Ruan Qiuqi Jiang Mingyan Institute of Information Sceince Northern Jiaotong University State Key Lab of Intelligent Tech & Systems Dept.of ComputerTsinghua University Dept of Electronic Shandong University

This paper presents the hybrid algorithm of global optimization of dynamic learning rate for multilayer feedforward neural networks(MLFNN). The effect of inexact line search on conjugacy was studied and a generalized conjugate gradient method based on this effect was proposed and shown to have global convergence for error backpagation of *** descent property and global convergence was given for the improved hybrid algrithm of conjugate gradient algorithm, the results of the proposed algorithm show a considerable improvement over the FletcherRreeves algorithm and conventional BP algorithm, it overcomes the drawback of conventional BP and Polak-Ribieve conjugate gradient algorithm that maybe plung into local minima.

关键词： the conjugate gradient algorithm global convergence backpropagation inexact line search

来源：评论

学校读者我要写书评

暂无评论

A novel fast learning algorithms for time-delay neural networks

A novel fast learning algorithms for time-delay neural netwo...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Jiang Minghu Zhu Xiaoyan The State Key Lab of Intelligent Tech. & Systems Department of Computer Tsinghua University Beijing China

To counter the drawbacks of long training time required by Waibel's time-delay neural networks (TDNN) in phoneme recognition, the paper puts forward several improved fast learning methods for TDNN. Merging the unsupervised Oja rule and the similar error backpropagation algorithm for initial training of TDNN weights can effectively increase the convergence speed. Improving the error energy function and updating the changing of weights according to size of output error, can increase the training speed. From backpropagation along layer, to average overlap part of backpropagation error of the first hidden layer along a frame, the training samples gradually increase the convergence speed increases. For multi-class phonemic modular TDNNs, we improve the architecture of Waibel's modular networks, and obtain an optimum modular TDNNs of tree structure to accelerate its learning. Its training time is less than Waibel's modular TDNNs.

关键词： Neural networks Convergence IEEE members intelligent systems intelligent networks Computer networks Counting circuits Learning systems Merging Computer errors

来源：评论

学校读者我要写书评

暂无评论

A neural network tracking controller for robot manipulators with unknown dynamics

引用

IFAC Proceedings Volumes 1999年第2期32卷 719-724页

作者： Fuchun Sun Zengqi Sun Chundi Mu Rongjun Zhang Department of Computer Sci. State Key Lab of Intelligent Tech. & Systems Tsinghua University Beijing 100084 P.R.China Department of Automation Tsinghua University Beijing 100084 P.R.China Department of Electrical Engineering Purdue University at Indianapolis Indianapolis IN 46202-5132 U.S.A.

A neural network (NN)-based adaptive control law is proposed for the tracking control of an n-link robot manipulator with unknown dynamic nonlinearities. Basis function-like networks are employed to approximate the plant nonlinearities, and the bound on the NN reconstruction error is assumed to be unknown. The proposed NN-based adaptive control approach integrates the NN approach and an adaptive discrete variable structure control with a simple estimation mechanism for the upper bound on the NN reconstruction errors, and additional control input as a function of the estimate. Lyapunov stability theory is used to prove the uniform ultimate boundedness of the tracking error.

关键词： robot adaptive control neural networks stability discrete variable structure

来源：评论

学校读者我要写书评

暂无评论

Neural adaptive controller design of robotic manipulators with an observer

引用

IFAC Proceedings Volumes 1999年第2期32卷 5485-5490页

作者： Fuchun Sun Zengqi Sun PengYung Woo Zhengdong Zhang Department of Computer Sci. State Key Lab of Intelligent Tech. & Systems Tsinghua University Beijing 100084 P. R.China. Department of Electrical Engineering Northern Illinois University Dekalb IL 60115 U.S.A. Department of Electric and Automatic Engineering Houbei University of Technology Tianjtn 300130 P. R. China.

A neural network(NN)-based adaptive controller with an observer is proposed in this paper for trajectory tracking of robotic manipulators with unknown dynamic nonlinearities. It is assumed that the robotic manipulator has only joint angle position measurements. The NN is used to approximate the revised robot dynamics function. The adaptive controller with an observer can guarantee the uniform ultimate bounds of the tracking errors and the observer errors as well as the bounds of the NN weights. Main theoretical results for designing such a controller are given, and the control performance of the proposed controller is verified with simulation studies.

关键词： Robot observer neural networks adaptive control stability

来源：评论

学校读者我要写书评

暂无评论

Computer Processing of Oriental Languages. Beyond the Orient: The Research Challenges Ahead 1

引用

丛书名： Lecture Notes in Computer Science

1000年

作者： Yuji Matsumoto Richard W. Sproat Kam-Fai Wong Min Zhang

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：