检索结果-内蒙古大学图书馆

arXiv 2017年

作者： Zhao, Kui Wang, Yi Hu, Xia Wang, Can Zhejiang Provincial Key Laboratory of Service Robot College of Computer Science and Technology Zhejiang University Hangzhou310027 China Hangzhou S&T Information Research Institute Hangzhou310014 China

Blog is becoming an increasingly popular media for information publishing. Besides the main content, most of blog pages nowadays also contain noisy information such as advertisements etc. Removing these unrelated elements can improves user experience, but also can better adapt the content to various devices such as mobile phones. Though template-based extractors are highly accurate, they may incur expensive cost in that a large number of template need to be developed and they will fail once the template is updated. To address these issues, we present a novel template-independent content extractor for blog pages. First, we convert a blog page into a DOM-Tree, where all elements including the title and body blocks in a page correspond to subtrees. Then we construct subtree candidate set for the title and the body blocks respectively, and extract both spatial and content features for elements contained in the subtree. SVM classifiers for the title and the body blocks are trained using these features. Finally, the classifiers are used to extract the main content from blog pages. We test our extractor on 2,250 blog pages crawled from nine blog sites with obviously different styles and templates. Experimental results verify the effectiveness of our extractor. Copyright © 2017, The Authors. All rights reserved.

关键词： Blogs

来源：评论

学校读者我要写书评

暂无评论

TripletGAN: Training generative model with triplet loss

arXiv

引用

arXiv 2017年

作者： Cao, Gongze Yang, Yezhou Lei, Jie Jin, Cheng Liu, Yang Song, Mingli Zhejiang Provincial Key Laboratory of Service Robot Zhejiang University School of Computing Informatics and Decision Systems Engineering Arizona State University School of Computer Science Fudan University Shanghai China School of Mathematics Zhejiang University Shanghai Alibaba Group Hang Zhou

As an effective way of metric learning, triplet loss has been widely used in many deep learning tasks, including face recognition and person-ReID, leading to many states of the arts. The main innovation of triplet loss is using feature map to replace softmax in the classification task. Inspired by this concept, we propose here a new adversarial modeling method by substituting the classification loss of discriminator to triplet loss. Theoretical proof based on IPM (Integral probability metric) demonstrates that such setting will help generator converge to the given distribution theoretically under some conditions. Moreover, since triplet loss requires the generator to maximize distance within a class, we justify tripletGAN is also helpful to prevent mode collapse through both theory and experiment. Copyright © 2017, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

An oriented random walk model for simulation test in the indoor navigation for visually impaired users

An oriented random walk model for simulation test in the ind...

引用

International Conference on Network Security and Communication Engineering, NSCE 2014

作者： Yu, Z. Zhang, X.B. Huang, K. Wang, W. Zhejiang Provincial Key Laboratory of Service Robot College of Computer Science Zhejiang University Hangzhou China

ISBN: (纸本)9781138028210

Position Detection is an important application in Wireless Sensor Network and has received much research attention. Recently, it has been used for indoor navigation for people with visual impairment. However, experiment and evaluation of these systems usually incur a high cost in that users with disability must be engaged. To address this issue, this paper proposes an Oriented Random Walk Model for the simulation test before sensors are deployed. Cost can be reduced and more efficient deployment can be deployed. © 2015 Taylor & Francis Group, London.

关键词： Wireless sensor networks

来源：评论

学校读者我要写书评

暂无评论

Paragraph reconstruction for postscript documents with complex layout

Paragraph reconstruction for postscript documents with compl...

引用

International Conference on Network Security and Communication Engineering, NSCE 2014

作者： Zhang, M.N. Li, M.Y. Wang, W. Yu, Z. Zhejiang Provincial Key Laboratory of Service Robot College of Computer Science Zhejiang University Hangzhou China

ISBN: (纸本)9781138028210

Electronic publishing has become an irreversible trend in publishing industry. An important step in electronic publishing is to convert a scanned copy or a digital image of a document into editable texts, which is very important to the people with visual impairment, who access the information by means of screen-readers. Existing methods for document parsing and paragraph reconstruction only work on PostScript documents with a simple and standard layout. These techniques are severely limited when applied to documents with complex layout and irregular typesetting. To address this issue, we propose a rectangle segmenting approach to make paragraph reconstruction for Postscript documents with complex layout. © 2015 Taylor & Francis Group, London.

关键词： Electronic publishing

来源：评论

学校读者我要写书评

暂无评论

Reduce the shopping distance: Map region search based on high order voronoi diagram

Reduce the shopping distance: Map region search based on hig...

引用

International Conference of Young computer Scientists, Engineers and Educators, ICYCSEE 2015

作者： Yu, Zhi Wang, Can Bu, Jiajun Zhang, Mengni Wu, Zejun Chen, Chun Zhejiang Provincial Key Laboratory of Service Robot College of Computer Science Zhejiang University Hangzhou310027 China

ISBN: (纸本)9783662462478

Many people would like to purchase items using locationbased services to find the suitable stores in daily life. Although there are many online map search engines giving isolated Point-of-Interest as query results according to the correlation between isolated stores and the query, this interaction is difficult in meeting the shopping needs of people with disabilities, who would usually prefer shopping in one single location to avoid inconvenience in transportation. In this article, we propose a framework of map search service using Region-of-Interest (ROI) as the query result, which can greatly reduce users shopping distance among multiple stores. High order Voronoi diagram is used to reduce the time complexity of Region-of-Interests generation. Experimental results show that our method is both efficient and effective. © Springer-Verlag Berlin Heidelberg 2015.

关键词： Search engines

来源：评论

学校读者我要写书评

暂无评论

An optimal sampling method for Web Accessibility Quantitative Metric 12

An optimal sampling method for Web Accessibility Quantitativ...

引用

12th International Web for All Conference, W4A 2015

作者： Zhang, Mengni Wang, Can Bu, Jiajun Yu, Zhi Lu, Yi Zhang, Ruijie Chen, Chun Zhejiang Provincial Key Laboratory of Service Robot College of Computer Science Zhejiang University Hangzhou310027 China

ISBN: (纸本)9781450333429

When evaluating the accessibility of a website, we usually resort to sampling methods to reduce the cost of evaluation. In this kind of approaches, a small subset of pages in a website are chosen for evaluating the accessibility value of the whole website. Good sampling quality means the selected subset can represent well the accessibility level of the whole website, i.e. minimizing the accessibility evaluation difference between the whole site and the sampled subset. As existing studies show the accuracy of sampling methods depends heavily on the metric, we propose in this paper a specific sampling method OPS-WAQM that is optimized for Web Accessibility Quantitative Metric (WAQM). OPS-WAQM minimizes the sampling error by choosing the optimal sample numbers in different page depth layers. A greedy algorithm is proposed to approximately solve the optimization problem in an efficient way. We use a dataset of 20 websites, 365780 web pages to validate our method. Experimental results show that our sampling method is effective for web accessibility evaluation. Copyright 2015 ACM.

关键词： Websites

来源：评论

学校读者我要写书评

暂无评论

WebRTC based remote rehabilitation guidance system

WebRTC based remote rehabilitation guidance system

引用

International Conference on Network Security and Communication Engineering, NSCE 2014

作者： Shi, Q.H. Zheng, J. Wang, W. Wu, Y. Yan, Z.J. Zhejiang Provincial Key Laboratory of Service Robot College of Computer Science Zhejiang University Hangzhou China Zhejiang Yuying College of Vocational Technology Hangzhou China

ISBN: (纸本)9781138028210

Compared with the large number of disabled population, medical resource is scarce and partially distributed. With the development of Internet and information technology, more and more services can be provided through the network, not restricted by distance. In this context, this paper developed a P2P instant communication system based on the WebRTC technology and realized a remote rehabilitation guidance system. The goal of this system is to make sure that a user can receive remote rehabilitation guidance scheme through computer or mobile phone. The system has no need to install additional software and plug-ins. The open web rehabilitation guidance can be realized through a browser only. © 2015 Taylor & Francis Group, London.

关键词： Remote control

来源：评论

学校读者我要写书评

暂无评论

Reduce the Shopping Distance:Map Region Search Based on High Order Voronoi Diagram

Reduce the Shopping Distance:Map Region Search Based on High...

引用

International Conference of Young computer Scientists, Engineers and Educators, ICYCSEE 2015

作者： Zhi Yu Can Wang Jiajun Bu Mengni Zhang Zejun Wu Chun Chen Zhejiang Provincial Key Laboratory of Service Robot College of Computer ScienceZhejiang University

Many people would like to purchase items using locationbased services to find the suitable stores in daily *** there are many online map search engines giving isolated Point-of-Interest as query results according to the correlation between isolated stores and the query,this interaction is difficult in meeting the shopping needs of people with disabilities,who would usually prefer shopping in one single location to avoid inconvenience in *** this article,we propose a framework of map search service using Region-of-Interest(ROI) as the query result,which can greatly reduce users shopping distance among multiple *** order Voronoi diagram is used to reduce the time complexity of Region-of-Interests *** results show that our method is both efficient and effective.

关键词： Map Search Region Search Voronoi Diagram

来源：评论

学校读者我要写书评

暂无评论

A Location-Based Personalized Focused Search System for Physically Challenged People

A Location-Based Personalized Focused Search System for Phys...

引用

2015 International Industrial Informatics and computer Engineering Conference(IIICEC 2015)

作者： Xin Shen Wei Wang Junyi Huang Yue Wu Zhejiang Provincial Key Laboratory of Service Robot College of Computer ScienceZhejiang University

Physically Challenged people are a special part of society,which need more *** of the huge differences in physical conditions,qualities of life,locations and personal backgrounds,there exists a lot of challenges for disabled people searching related information and receive professional and personal *** paper proposes a novel focused search system for physically challenged people,which uses the location information and provides the more effective search *** search system consists of two function models,one is a focused search engine,and the other is a map of the related *** search engine is designed for the multimedia files of disabled recovery,including words,pictures,audios and *** map of related institutions offers the location-based service showing the near occasion addresses intuitively,where the physically challenged people may get professional *** this location-based focused search system,the special community has an easy access to the technical database and relevant *** system is highly significant in the rehabilitation information acquisition field.

关键词： Focused Search System Location-Based service Personalized Retrieval

来源：评论

学校读者我要写书评

暂无评论

Error Compensation Strategy for Parallel Error Diffusion

Error Compensation Strategy for Parallel Error Diffusion

引用

International Symposium on Computational Intelligence and Design, ISCID

作者： Wenhua Zhu College of Computer Science Zhejiang Provincial Key Laboratory of Service Robot Hangzhou China

ISBN: (纸本)9781467395885

Parallelizing error diffusion is a difficult task because of its inherently serial character in the original form. In this paper, we describe an error compensation strategy for parallel error diffusion. In our method, original image is split into independent blocks. In order to compensate for the missing errors around the boundary of blocks, overlapping areas and initial error of boundary pixels are all introduced. Experimental results show that the proposed method generates high quality images similar to that of the original algorithm.

关键词： Error compensation Quantization (signal) Modulation Parallel processing Partitioning algorithms Program processors Printers

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：