检索结果-内蒙古大学图书馆

iCatse International Conference on Information Science and Applications (ICISA)

作者： Shen, Ying Zhang, Qiang Zhang, Jin Huang, Jiyue Lu, Yuming Lei, Kai Peking Univ Sch Elect & Comp Engn Shenzhen Key Lab Cloud Comp Technol & Applicat Inst Big Data Technol Shenzhen 518055 Peoples R China Shenzhen Inst Informat Technol ShenZhen Key Lab Visual Media Proc & Streaming Me Shenzhen 518172 Peoples R China

ISBN: (纸本)9789811310560;9789811310553

Automatic text classification (TC) research can be used for real-world problems such as the classification of in-patient discharge summaries and medical text reports, which is beneficial to make medical documents more understandable to doctors. However, in electronic medical records (EMR), the texts containing sentences are shorter than that in general domain, which leads to the lack of semantic features and the ambiguity of semantic. To tackle this challenge, we propose to add word-cluster embedding to deep neural network for improving short text classification. Concretely, we first use hierarchical agglomerative clustering to cluster the word vectors in the semantic space. Then we calculate the cluster center vector which represents the implicit topic information of words in the cluster. Finally, we expand word vector with cluster center vector, and implement classifiers using CNN and LSTM respectively. To evaluate the performance of our proposed method, we conduct experiments on public data sets TREC and the medical short sentences data sets which is constructed and released by us. The experimental results demonstrate that our proposed method outperforms state-of-the-art baselines in short sentence classification on both medical domain and general domain.

关键词： Text classification Word-cluster embedding Hierarchical agglomerative clustering

来源：评论

学校读者我要写书评

暂无评论

Information Centric Networking media streaming Experiment Platform Design 2nd

Information Centric Networking Media Streaming Experiment Pl...

引用

2nd International Conference on Smart Computing and Communication (SmartCom)

作者： Lu, Yuming Hu, Tao Wang, Xiaojun Shenzhen Inst Informat Technol Shenzhen Key Lab Visual Media Proc & Streaming Me Shenzhen 518172 Peoples R China

ISBN: (纸本)9783319738307;9783319738291

Information Centric Networking (ICN) is a revolutionary concept that considers data content interconnected networking instead of equipment interconnected data transmission. Designing and optimizing such a system require not only a software simulator environment but also an experimental platform to test new type of big-data processing architecture. This paper introduce and contribute on put forward the idea of a new distributed publish-subscribe media big data processing platforms base on the ICN architecture. First part discusses the component that included in a ICN media streaming platform, and then describes the design of a distributed publish-subscribe (P/S) system, which manages the subscription query, storage, transmission and content scheduling that complies with the named data networking (NDN) framework. The simulation utilizes software defined network (SDN) controller to carry out artificial intelligence supported computing in the cloud to generate global view of the distributed media content.

关键词： streaming media Information center networking Publish-subscribe systems Software defined networks Named data networks

来源：评论

学校读者我要写书评

暂无评论

MSA vs. MVC: Future Trends for Big Data processing Platforms 2nd

MSA vs. MVC: Future Trends for Big Data Processing Platforms

引用

2nd International Conference on Smart Computing and Communication (SmartCom)

作者： Lu, Yuming Liu, Wei Cui, Haoxiang Shenzhen Inst Informat Technol Shenzhen Key Lab Visual Media Proc & Streaming Me Shenzhen 518172 Peoples R China Shenzhen Univ Coll Informat Engn Shenzhen 518060 Peoples R China

ISBN: (纸本)9783319738307;9783319738291

Big data processing systems design is highly prioritized concern for both academia and industry. The conventional MVC architecture exposes limitations on system scalability and consistency. The task of integrating new services into an existing commercial application platform has become an impossible task and torturing nightmare for the system development team. The innovative MSA architecture is aimed to solve such a problem. The main contribution of this paper is comparison between the MSA and MVC system design and development architectures, summaries future research and development issues and challenges. This paper first discusses the problems and challenges of big data management, compares and discusses the characteristics of MVC and MSA patterned big data processing (BDP) platforms. Then we verify the MSA big data management systems, distributed data storage and the progress of the large data storage architecture utilize an experimental BDP platform. Finally list future research and development direction to provide valuable reference for further work.

关键词： Micro-service architecture Big data processing platform

来源：评论

学校读者我要写书评

暂无评论

ICNRL: An Initiative Framework Towards Information Centric Network Representation 1

ICNRL: An Initiative Framework Towards Information Centric N...

引用

1st IEEE International Conference on Hot Information-Centric Networking (HotICN)

作者： Lu, Yuming Li, Weichao Wang, Xiaojun ShenZhen Inst Informat Technol ShenZhen Key Lab Visual Media Proc & Streaming Me Shenzhen Peoples R China Southern Univ Sci & Technol Inst Future Networks Shenzhen Peoples R China

ISBN: (纸本)9781538648704

The exponentially growing demand for computational resources prevents the Information Centric Networking (ICN) being deployed in practice due to the high dimensional sparse data computation. However, we argue that Network Representation Learning (NRL) can help to solve the problem by transforming the raw network information data into low-dimensional dense adjacency matrix representation. In this paper, we propose ICNRL, a novel task-based NRL scheme for ICN. Based on the adjacency matrix generated by NRL, ICNRL can calculate the index threshold value to support the networking decision making of content-store (CS), pending information table (PIT), and forwarding information base (FIB), and therefore improves the management and processing capabilities of ICN.

关键词： Task analysis Matrix converters Sparse matrices Indexes Decision making Routing media

来源：评论

学校读者我要写书评

暂无评论

Infrared-Visible Image Fusion Based on Convolutional Neural Networks (CNN) 8th

Infrared-Visible Image Fusion Based on Convolutional Neural ...

引用

8th International Conference on Intelligence Science and Big Data Engineering (IScIDE)

作者： Ren, Xianyi Meng, Fanyang Hu, Tao Liu, Zhijun Wang, Changwei Shenzhen Inst Informat Technol Shenzhen Peoples R China Shenzhen Key Lab Visual Media Proc & Streaming Me Shenzhen Peoples R China Intelligent Vis Engn Res Ctr Guangdong Prov Guangzhou Peoples R China

ISBN: (纸本)9783030026981;9783030026974

Image fusion is a process of combing multiple images of the same scene into a single image with the aim of preserving the full content information and retaining the important features from each of the original images. In this paper, a novel image fusion method based on Convolutional Neural Networks (CNN) and saliency detection is proposed. Here, we use the image representations derived from CNN Network optimized for infrared-visible image fusion. Since the lower layers of the network can seize the exact value of the original image, and the high layers of the network can capture the high-level content in terms of objects and their arrangement in the input image, we exploit more low-layer features of visible image and more high-layer features of infrared image in the fusion. And during the fusion procedure, the infrared target of an infrared image is effectively highlighted using saliency detection method and only the salient information of the infrared image will be fused. The method aimed to preserve the abundant detail information from visible image as much as possible, meanwhile preserve the salient information in the infrared image. Experimental results show that the proposed fusion method is rather promising.

关键词： Image fusion Convolutional Neural Networks (CNN) Saliency detection

来源：评论

学校读者我要写书评

暂无评论

Improving Medical Short Text Classification with Semantic Expansion Using Word-Cluster Embedding

arXiv

引用

arXiv 2018年

作者： Shen, Ying Zhang, Qiang Zhang, Jin Huang, Jiyue Lu, Yuming Lei, Kai Peking University SHENZHEN518055 China ShenZhen Key Lab for Visual Media Processing and Streaming Media ShenZhen Institute of Information Technology SHENZHEN518172 China

. Automatic text classification (TC) research can be used for real-world problems such as the classification of in-patient discharge summaries and medical text reports, which is beneficial to make medical documents more understandable to doctors. However, in electronic medical records (EMR), the texts containing sentences are shorter than that in general domain, which leads to the lack of semantic features and the ambiguity of semantic. To tackle this challenge, we propose to add word-cluster embedding to deep neural network for improving short text classification. Concretely, we first use hierarchical agglomerative clustering to cluster the word vectors in the semantic space. Then we calculate the cluster center vector which represents the implicit topic information of words in the cluster. Finally, we expand word vector with cluster center vector, and implement classifiers using CNN and LSTM respectively. To evaluate the performance of our proposed method, we conduct experiments on public data sets TREC and the medical short sentences data sets which is constructed and released by us. The experimental results demonstrate that our proposed method outperforms state-of-the-art baselines in short sentence classification on both medical domain and general domain. Copyright © 2018, The Authors. All rights reserved.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

PSVA: A Content-Based Publish/Subscribe Video Advertising Framework 2nd

PSVA: A Content-Based Publish/Subscribe Video Advertising Fr...

引用

2nd International Conference on Smart Computing and Communication (SmartCom)

作者： Wang, Feiyang Zhang, Dongyu Lu, Yuming Lei, Kai Peking Univ Inst Big Data Technol Shenzhen Key Lab Cloud Comp Technol & Applicat Sch Elect & Comp Engn SECE Shenzhen 518055 Peoples R China Harbin Inst Technol Harbin Peoples R China Shenzhen Inst Informat Technol Shenzhen Key Lab Visual Media Proc & Streaming Me Shenzhen 518172 Peoples R China

ISBN: (纸本)9783319738307;9783319738291

Online video technology rapidly developed in the past few decades, amounts all applications, video advertising is a uprising approach to improve commercial advertisement revenue due to its diverse presentations, interactivity GUI, tailored media publishing and measurable subscription features etc. However to accurately match commercial adverts from the publisher's media source, with consumer desire contents, as well as effective real-time video streaming transmission in a media processing system remains a problem. This paper proposes a content-based Publish/Subscribe Video Advertising (PSVA) framework. In this framework, we construct a high-performance scalable communication subsystem to carry out real-time video streaming transmission, and the handle event matching by a stream computation subsystem. The experimental results show that the PSVA framework is effective.

关键词： Online video Publish/subscribe system Video advertising Event matching

来源：评论

学校读者我要写书评

暂无评论

Design and Implementation of E-government Oriented Rapid BPM OA Platform 4

Design and Implementation of E-government Oriented Rapid BPM...

引用

4th International Conference on Systems and Informatics (ICSAI)

作者： Wang, Yukun Lei, Kai Wu, Jincheng Huang, Xiaohui Lu, Yuming Peking Univ Inst Big Data Technol Shenzhen Key Lab Cloud Comp Technol & Applicat Sch Elect & Comp Engn Shenzhen 518055 Peoples R China Natl Supercomp Ctr Shenzhen Shenzhen Cloud Comp Ctr Shenzhen 518055 Peoples R China Shenzhen Inst Informat Technol Shenzhen Key Lab Cognit Serv Engn Shenzhen 518172 Peoples R China Shenzhen Inst Informat Technol Shenzhen Key Lab Streaming Media & Visual Media P Shenzhen 518172 Peoples R China

ISBN: (纸本)9781538611074

The e-government system plays a prominent part in government's office work, however, in the development process, there exist some problems currently, such as repeated development of common functions but hard to satisfy various personalized needs, independence of different systems which makes information difficult to share. These problems have seriously hindered development progress, increased development cost, and may result in information isolated islands. To solve these problems, this paper proposes an e-government oriented BPM OA platform, and introduces its design and implementation process. To achieve rapid development, by analysis and abstraction, we extract main features of common e-government systems and make them into configurable functions. The powerful form builder, the explicit authority management method and the BPMN 2.0 based workflow engine can significantly simplify the development and deployment process, promote the inter-system information exchange, and improve development efficiency.

关键词： BPM e-government rapid development work flow

来源：评论

学校读者我要写书评

暂无评论

A novel simple candidate set method for symmetric TSP and its application in MAX-MIN ant system

A novel simple candidate set method for symmetric TSP and it...

引用

3rd International Conference on Swarm Intelligence, ICSI 2012

作者： Deng, Miao Zhang, Jihong Liang, Yongsheng Lin, Guangming Liu, Wei School of Information Engineering Shenzhen University Shenzhen China Shenzhen Key Lab. of Visual Media Processing and Transmission Shenzhen Institute of Information Technology Shenzhen China

ISBN: (纸本)9783642309755

Traveling Salesman Problem (TSP) is a kind of typical NP problem and has been extensively researched in combinatorial optimization. For solving it more effectively, candidate set is used in many algorithms in order to limit the selecting range when choosing next city to move, such as in Ant Systems, or to initialize a local optimum solution, such as in Lin-Kernighan Heuristic (LKH) algorithm. A novel simple method for generating candidate set is proposed in this paper and applied into MAX-MIN Ant System (MMAS) for symmetric TSP problem. Experimental results show that it has better performance than other Ant Systems including MMAS. Moreover, this method can be used in other algorithms for symmetric TSP problem. © 2012 Springer-Verlag.

关键词： Traveling salesman problem

来源：评论

学校读者我要写书评

暂无评论

A Discrete-time Switching Neural Network for Quadratic Programming

A Discrete-time Switching Neural Network for Quadratic Progr...

引用

International Joint Conference on Neural Networks

作者： S. Chen S. Li Y. Liang Y. Lou Key Lab of Visual Media Processing and Transmission Shenzhen Institute of Information Technology Stvens Institute of Technology School of Mechatronics and Information Yiwu Industrial and Commercial College

ISBN: (纸本)9781467314886

This paper presents a discrete-time neural network with a switching structure to solve a general quadratic programming problem in real time. Compared with existing ones for solving quadratic programming problems, the proposed neural network model has a simple architecture and uses a limited number of neurons to solve the problem, irrespective of the dimension of the decision variables or the number of constraints. The global convergence of the model is proven using contraction theory. Simulations are performed to demonstrate the effectiveness of the proposed method.

关键词： Neural network Quadratic programming Contraction theory Global convergence quadratic programming Neural network discrete-time dimensions switching

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：