检索结果-内蒙古大学图书馆

7th International Symposium on Visual Information Communication and Interaction, VINCI 2014

作者： Guo, Xiaofei Li, Juanzi Yang, Ruibing Ma, Xiaoli Knowledge Engineering Group Department of Computer Science and Technology Tsinghua University 10084 China

ISBN: (纸本)9781450327657

Nowadays, there are many events reported by News Media everyday, which contains a massive number of news. People are getting more and more interested in understanding how an event evolves after it happens. News related to the same event or similar events usually has more common entities and stronger topic correlations, which is a new perspective to study news event. Due to the complexity of event evolving process, event visualization has been a big challenge for a long time. In this paper, we design a novel four-phase framework NEI(News Event Insight) that focuses on visualizing a news event properly and clearly, namely (1)Entity Topic Modeling. We extract topics and entities through timeline. (2)Temporal Topic Correlation Analysis. Based on the topic modeling result, we design two methods to select hot topics and build links for them. (3)Keyword Extraction. Specially, we combine string frequency with syntax features and use language models to acquire candidate keywords for representing topics. (4)Visualization. Visualization demonstrates the quantifying properties of topics related to a certain event. A case study shows our framework achieves promising results on both single event and similar events. Copyright 2014 ACM.

关键词： Visualization

来源：评论

学校读者我要写书评

暂无评论

Sentential paraphrase generation for agglutinative languages using SVM with a string kernel 28

Sentential paraphrase generation for agglutinative languages...

引用

28th Pacific Asia Conference on Language, Information and Computation, PACLIC 2014

作者： Park, Hancheol Gweon, Gahgene Choi, Ho-Jin Heo, Jeong Ryu, Pum-Mo Department of Knowledge Service Engineering KAIST Japan Department of Computer Science KAIST Japan Knowledge Mining Research Team ETRI Japan

ISBN: (纸本)9786165518871

Paraphrase generation is widely used for various natural language processing (NLP) applications such as question answering, multi-document summarization, and machine translation. In this study, we identify the problems occurring in the process of applying existing probabilistic model-based methods to agglutinative languages, and provide solutions by reflecting the inherent characteristics of agglutinative languages. More specifically, we propose and evaluate a sentential paraphrase generation (SPG) method for the Korean language using Support Vector Machines (SVM) with a string kernel. The quality of generated paraphrases is evaluated using three criteria: (1) meaning preservation, (2) grammaticality, and (3) equivalence. Our experiment shows that the proposed method outperformed a probabilistic model-based method by 12%, 16%, and 17%, respectively, with respect to the three criteria. Copyright 2014 by Hancheol Park, Gahgene Gweon, Ho-Jin Choi, Jeong Heo, and Pum-Mo Ryu.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Adaptive tourism modeling and socialization system

Adaptive tourism modeling and socialization system

引用

2009 IEEE International Conference on Social Computing, SocialCom 2009

作者： Coelho, Bruno Martins, Constantino Almeida, Ana GECAD - Knowledge Engineering and Decision Support Group Computer Science Department Institute of Engineering Porto Portugal

ISBN: (纸本)9780769538235

This paper presents an undertaken research work about the development of an Adaptive Tourism Modeling System which attempts to correctly model a tourism web application user profile. This paper will follow the methodology used behind the concept of the application, 'User Modeling as a process', which consists of three specific modules in order to achieve a clear user modeling technology. The user representation module devises a collection of knowledge representation formalisms that co-exist in order to constitute the system's view of the user;the user reasoning module is responsible for the execution of several inference tasks as well as the generation of possibly new and valuable information about the user, using the latter module. The user concept that comes out of these two modules will ultimately be used by various kinds of application-level systems, namely a recommender system, therefore instantiating a project-dependent third module. © 2009 IEEE.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Certificate management system using blockchain 11th

Certificate management system using blockchain

引用

11th International Conference on Soft Computing and Pattern Recognition, SoCPaR 2019, and 11th World Congress on Nature and Biologically Inspired Computing, NaBIC 2019

作者： Endurthi, Anjaneyulu Khare, Akhil Computer Science and Engineering Department Rajiv Gandhi University of Knowledge Technologies BasarTelangana India Computer Science and Engineering Department MVSR Engineering College HyderabadTelangana India

ISBN: (纸本)9783030493448

Overcoming the problems of existing file storage and sharing approaches, this paper gives an efficient way of storing and sharing the files using blockchain and smart contracts. As an example, Managing student’s information and sharing them or giving access to unauthorized people in a much secured way is one of the major responsibilities of any university or organization. Instead of storing the certificates in a normal database, blockchain technology can be used to store the certificates. In our approach, the certificates and other documents will be uploaded to the blockchain and will be shared securely when any third party needs to access the certificates. The university administrator has ultimate rights to upload student’s certificates and managing them. Students can view their documents and certificates;other unauthorized people (third party/organization/employer/others) can access the files after taking permission of the college administrator. These types of approaches are very useful in this era where efficient storage and sharing the files plays a crucial role. The feature which differentiates our approach of certificate management system from existing certificate management system is, our system is decentralized, cryptographically secure, immutable and efficient. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021.

关键词： Smart contract

来源：评论

学校读者我要写书评

暂无评论

Security Issues and Defensive Approaches in Deep Learning Frameworks

引用

Tsinghua science and Technology 2021年第6期26卷 894-905页

作者： Hongsong Chen Yongpeng Zhang Yongrui Cao Jing Xie Department of Computer Science University of Science and Technology Beijing(USTB)Beijing 100083China Beijing Key Laboratory of Knowledge Engineering for Materials Science Beijing 100083China Department of Computer Science and Technology University of Science and Technology Beijing(USTB)Beijing 100083China Defense Electronics Institute China Industrial Control System Cyber Emergency Response TeamBeijing 100040China

Deep learning frameworks promote the development of artificial intelligence and demonstrate considerable potential in numerous ***,the security issues of deep learning frameworks are among the main risks preventing the wide application of *** on deep learning frameworks by malicious internal or external attackers would exert substantial effects on society and *** start with a description of the framework of deep learning algorithms and a detailed analysis of attacks and vulnerabilities in *** propose a highly comprehensive classification approach for security issues and defensive approaches in deep learning frameworks and connect different attacks to corresponding defensive ***,we analyze a case of the physical-world use of deep learning security *** addition,we discuss future directions and open issues in deep learning *** hope that our research will inspire future developments and draw attention from academic and industrial domains to the security of deep learning frameworks.

关键词： adversarial examples deep learning frameworks defensive approaches security issues

来源：评论

学校读者我要写书评

暂无评论

ExerLink: Enabling pervasive social exergames with heterogeneous exercise devices 12

ExerLink: Enabling pervasive social exergames with heterogen...

引用

10th International Conference on Mobile Systems, Applications, and Services, MobiSys'12

作者： Park, Taiwoo Hwang, Inseok Lee, Uichin Lee, Sunghoon Ivan Yoo, Chungkuk Lee, Youngki Jang, Hyukjae Choe, Sungwon Peter Park, Souneil Song, Junehwa Department of Computer Science KAIST Korea Republic of Department of Knowledge Service Engineering KAIST Korea Republic of Computer Science Department UCLA United States

ISBN: (纸本)9781450313018

We envision that diverse social exercising games, or exergames, will emerge, featuring much richer interactivity with immersive game play experiences. Further, the recent advances of mobile devices and wireless networking will make such social engagement more pervasive - people carry portable exergame devices (e.g., jump ropes) and interact with remote users anytime, anywhere. Towards this goal, we explore the potential of using heterogeneous exercise devices as game controllers for a multi-player social exergame;e.g., playing a boat paddling game with two remote exercisers (one with a jump rope, and the other with a treadmill). In this paper, we propose a novel platform called ExerLink that converts exercise intensity to game inputs and intelligently balances intensity/delay variations for fair game play experiences. We report the design considerations and guidelines obtained from the design and development processes of game controllers. We validate the efficacy of game controllers and demonstrate the feasibility of social exergames with heterogeneous exercise devices via extensive human subject studies. © 2012 ACM.

关键词： Rope

来源：评论

学校读者我要写书评

暂无评论

Accumulated gradient normalization 9

Accumulated gradient normalization

引用

9th Asian Conference on Machine Learning, ACML 2017

作者： Hermans, Joeri R. Spanakis, Gerasimos Möckel, Rico Department of Electrical Engineering and Computer Science Liège University Belgium Department of Data Science and Knowledge Engineering Maastricht University Netherlands

This work addresses the instability in asynchronous data parallel optimization. It does so by introducing a novel distributed optimizer which is able to efficiently optimize a centralized model under communication constraints. The optimizer achieves this by pushing a normalized sequence of first-order gradients to a parameter server. This implies that the magnitude of a worker delta is smaller compared to an accumulated gradient, and provides a better direction towards a minimum compared to first-order gradients, which in turn also forces possible implicit momentum fluctuations to be more aligned since we make the assumption that all workers contribute towards a single minima. As a result, our approach mitigates the parameter staleness problem more effectively since staleness in asynchrony induces (implicit) momentum, and achieves a better convergence rate compared to other optimizers such as asynchronous easgd and dynsgd, which we show empirically. © 2017 J.R. Hermans, G. Spanakis & R. Möckel.

关键词： Optimization

来源：评论

学校读者我要写书评

暂无评论

Collective fast delivery by energy-efficient agents 43

Collective fast delivery by energy-efficient agents

引用

43rd International Symposium on Mathematical Foundations of computer science, MFCS 2018

作者： Bärtschi, Andreas Graf, Daniel Mihalák, Matúš Department of Computer Science ETH Zürich Switzerland Department of Data Science and Knowledge Engineering Maastricht University Netherlands

ISBN: (纸本)9783959770866

We consider k mobile agents initially located at distinct nodes of an undirected graph (on n nodes, with edge lengths). The agents have to deliver a single item from a given source node s to a given target node t. The agents can move along the edges of the graph, starting at time 0, with respect to the following: Each agent i has a weight ωi that defines the rate of energy consumption while travelling a distance in the graph, and a velocity υi with which it can move. We are interested in schedules (operating the k agents) that result in a small delivery time T (time when the item arrives at t), and small total energy consumption E. Concretely, we ask for a schedule that: either (i) Minimizes T , (ii) Minimizes lexicographically (T, E) (prioritizing fast delivery), or (iii) Minimizes · T + (1 − ) · E, for a given ∈ (0, 1). We show that (i) is solvable in polynomial time, and show that (ii) is polynomial-time solvable for uniform velocities and solvable in time O(n + k log k) for arbitrary velocities on paths, but in general is NP-hard even on planar graphs. As a corollary of our hardness result, (iii) is NP-hard, too. We show that there is a 2-approximation algorithm for (iii) using a single agent. © Andreas Bärtschi, Daniel Graf, and Matúš Mihalák.

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

An Efficient Embedded Security System for Reduce Car Accident to Build Safer World Based on IoT

An Efficient Embedded Security System for Reduce Car Acciden...

引用

2019 International Conference on Artificial Intelligence and Data Processing Symposium, IDAP 2019

作者： Othman, Nashwan Adnan Aydin, Ilhan Karakose, Mehmet Computer Science Department College of Science Knowledge University Erbil Kurdistan Region Iraq Computer Engineering Department Firat University Elazig Turkey

ISBN: (纸本)9781728129327

Nowadays, people need to work hard to afford their life and this results in people to live under the stress of the modern life style. This pressure of the modern life style might cause problems and accidents, for example, when a driver feels drowsy while driving his car, there is a great possibility of making an accident. For the sake of solving or reducing the amount of such accidents, we designed an embedded system. This system presents an enhanced technique for detecting drowsiness in real-Time and reducing car accident. In this paper, we proposed an eye blinking technique for real-Time video by using machine learning and computer vision techniques. When a driver often feels drowsy, instead of stopping the car for a while to rest and gain back his/her conscious, he/she continues driving, so we utilize Internet of Things (IoT), which is a system connected to internet. When a driver feels sleepy, directly an alert message will be sent to one of his nearest kin or friends. Then the alerted person can call the driver and oblige him/her to stop the car. Meanwhile, there is a speaker inside the car to warn the driver. To implement this system, a jetson TX2 embedded platform is used and it uses its own camera board for taking frames. © 2019 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Reinforcement learning transfer via common subspaces

Reinforcement learning transfer via common subspaces

引用

2011 Adaptive and Learning Agents Workshop, ALA 2011, Held at Autonomous Agents and Multi-Agent Systems Conference, AAMAS 2011

作者： Ammar, Haitham Bou Taylor, Matthew E. Department of Knowledge Engineering Maastricht University Netherlands Department of Computer Science Lafayette College United States

ISBN: (纸本)9783642284984

Agents in reinforcement learning tasks may learn slowly in large or complex tasks - transfer learning is one technique to speed up learning by providing an informative prior. How to best enable transfer between tasks with different state representations and/or actions is currently an open question. This paper introduces the concept of a common task subspace, which is used to autonomously learn how two tasks are related. Experiments in two different nonlinear domains empirically show that a learned inter-state mapping can successfully be used by fitted value iteration, to (1) improving the performance of a policy learned with a fixed number of samples, and (2) reducing the time required to converge to a (near-) optimal policy with unlimited samples. © 2012 Springer-Verlag Berlin Heidelberg.

关键词： Reinforcement learning

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：