检索结果-内蒙古大学图书馆

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Xuechen Zhao Lei Tian Feng Xie Bin Zhou Haiyang Wang Hongzhou Wu Liqun Gao School of Computer National University of Defense Technology Changsha China Key Lab. of Software Engineering for Complex Systems National University of Defense Technology Changsha China

Zero-shot stance detection aims to determine the stance of previously unseen targets during the inference phase. Achieving effective feature alignment from seen targets to unseen targets is crucial for zero-shot stance detection. In this paper, we propose MSFR, a hierarchical contrastive learning framework, which consists of two core components: inter-aspect contrastive learning for distinguishing aspect-level features and intra-aspect contrastive learning for capturing attribute-level features. Specifically, inter-aspect contrastive learning first maps the global features of an utterance to multiple aspects that influence semantic expression (referred to as aspect-level feature differentiation). This process facilitates the alignment of semantic features across different factors of seen and unseen targets. Intra-aspect contrastive learning enhances the distinguishability of features within the same aspect (referred to as attribute-level feature differentiation) and improves the model’s fine-grained generalization capability. Experimental results demonstrate the superior performance of our model compared to competing baseline models.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Zero-Shot Stance Detection via Sentiment-Stance Contrastive Learning

Zero-Shot Stance Detection via Sentiment-Stance Contrastive ...

引用

International Conference on Tools for Artificial Intelligence (ICTAI)

作者： Jiaying Zou Xuechen Zhao Feng Xie Bin Zhou Zhong Zhang Lei Tian School of Computer National University of Defense Technology Changsha China Key Lab. of Software Engineering for Complex Systems National University of Defense Technology Changsha China

Zero-shot stance detection (ZSSD) is an important research problem that requires algorithms to have good stance detection capability even for unseen targets. In general, stance features can be grouped into two types: target-invariant and target-specific. Target-invariant features express the same stance regardless of the targets they are associated with, and such features are general and transferable. On the contrary, target-specific features will only be directly associated with specific targets. Therefore, it is crucial to effectively mine target-invariant features in texts in ZSSD. In this paper, we develop a method based on contrastive learning to mine certain transferable target-invariant expression features in texts from two dimensions of sentiment and stance and then generalize them to unseen targets. Specifically, we first grouped all texts into several types in terms of two orthogonal dimensions: sentiment polarity and stance polarity. Then we devise a supervised contrastive learning-based strategy to capture each type's common and transferable expressive features. Finally, we fuse the above-mentioned expressive features with the semantic features of the original texts about specific targets to deal with the stance detection for unseen targets. Extensive experiments on three benchmark datasets show that our proposed model achieves the state-of-the-art performance on most datasets. Code and other resources are availab.e on GitHub 1 1 https://***/zoujiaying1995/sscl-project.

关键词： Training Analytical models Codes Fuses Semantics Benchmark testing Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Multitask Learning Neural Networks for Pandemic Prediction with Public Stance Enhancement

Multitask Learning Neural Networks for Pandemic Prediction w...

引用

International Conference on Tools for Artificial Intelligence (ICTAI)

作者： Zhong Zhang Feng Xie Xuechen Zhao Bin Zhou Jian Chen Lei Tian School of Computer National University of Defence Technology Changsha China Key Lab. of Software Engineering for Complex Systems National University of Defence Technology Changsha China

State and local governments have imposed health policies to contain the spread of COVID-19 since it had a serious impact on human daily life. However, the public stance on these measures may be time-varying. It is likely to escalate the infection in the area where the public is negative or resistant. To take advantage of the correlation between public stance on health policies and the COVID-19 statistics, we propose a novel framework, Multitask Learning Neural Networks for Pandemic Prediction with Public Stance Enhancement (MP3), which is composed of three modules: (1) Stance awareness module to make stance detection on health policies from users' tweets in social media and convert them into a stance time series. (2) Temporal feature extraction module that applies Convolution Neural Network and Recurrent Neural Network to extract and fuse local patterns and long-term correlations from COVID-19 statistics. Moreover, a Stance Latency-aware Attention is proposed to capture dynamic social effects and fuse them with temporal features. (3) Multi-task prediction module to adopt Graph Convolution Network to model the spread of pandemic and employ multi-task learning to simultaneously predict COVID-19 statistics and the trend of public stance on health policies. The proposed framework outperforms state-of-the-art baselines on both confirmed cases and deaths prediction tasks.

关键词： COVID-19 Correlation Pandemics Social networking (online) Fuses Convolution Multitasking

来源：评论

学校读者我要写书评

暂无评论

Exploiting Crowd-based Learning Method in software engineering Course

Exploiting Crowd-based Learning Method in Software Engineeri...

引用

International Conference on Computer Science & Education (ICCSE)

作者： Xinjun Mao Yao Lu Yi Yang Key Lab. of Software Engineering for Complex Systems National University of Defense Technology Changsha China College of Computer National University of Defense Technology Changsha China

ISBN: (数字)9781728172675

ISBN: (纸本)9781728172682

software engineering is one of main major course for computer science. The main objective of this course is to improve the engineering capabilities of students to develop complex software systems. However, current course practice methods are teacher-centered and thus highly depend on the limited talents, experiences and efforts of teachers. Such methods greatly restrict students' scopes to gain necessary skills and artifacts, and to draw on successful experiences to address various personalized engineering issues and therefore enhance abilities to tackle engineering complexity. This paper presents a crowd-based learning method and introduces our exploration to use the method to reform the software engineering course practice. The essence of our exploration is to exploit the wisdoms and artifacts of crowds over Internet to support students to accomplish their course projects. The former is the engineering knowledges shared in question and answering (Q&A) platforms such as StackOverflow, and the latter is the open source software (OSS) hosted in OSS sites such as GitHub. We use the method in two separate course practices that aim to improve students' abilities to develop original, high-quality and large-scale software systems. The results of three-year explorations are positive and impressive and show that the crowd-based learning method is effective.

关键词： Education software engineering Learning systems Crowdsourcing software systems

来源：评论

学校读者我要写书评

暂无评论

Adaptive and impulsive cluster synchronization of a general complex dynamical network

Adaptive and impulsive cluster synchronization of a general ...

引用

2010 International Conference on Networking, Sensing and Control, ICNSC 2010

作者： Shi, Bingxiong Lu, Junan Lü, Jinhu Yu, Xinghuo College of Mathematics and Statistics State Key Lab. of Software Engineering Wuhan University Wuhan 430072 China Key Laboratory of Systems and Control Institute of Systems Science Chinese Academy of Sciences Beijing 100190 China School of Electrical and Computer Engineering RMIT University Melbourne 3000 Australia School of Electrical and Computer Engineering RMIT University Melbourne VIC 3001 Australia

ISBN: (纸本)9781424464531

Recently, cluster synchronization has attracted an increasing interest in various disciplines, includingmathematics, physics, and engineering sciences. This paper further investigates the cluster synchronization of a complex dynamical network by using adaptive and impulsive control techniques, respectively. Based on a complex dynamical network with two different clusters, several adaptive and impulsive cluster synchronization criteria are then proposed. It should be especially pointed out that the dynamics and its dimensions of the individual dynamical systems are different in the two different clusters. Also, some numerical simulations are given to verify the effectiveness of the above theoretical results. ©2010 IEEE.

关键词： Dynamical systems

来源：评论

学校读者我要写书评

暂无评论

A new method of illumination invariant face recognition

A new method of illumination invariant face recognition

引用

2nd International Conference on Innovative Computing, Information and Control, ICICIC 2007

作者： Wang, Juanjuan Wu, Lifang He, Xiaoguang Tian, Jie School of Electronic Information and Control Engineering Beijing University of Technology Beijing 100022 China Key Lab. of Complex Systems and Intelligence Science Institute of Automation Chinese Academy of Sciences P.O.Box 2728 Beijing 100080 China

ISBN: (纸本)0769528821

In this paper the Morphological Quotient Image (MQI) based on close and open operation are compared, and the more effective operation is chosen. Furthermore, the influence of template size to recognition rate is studied. For different size of face images, the optimal template is obtained by experiment using Yale Face Database B. The optimal template is tested using CMU-PIE database and the experimental result is encouraging. © 2007 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

A hybrid algorithm of facial feature point location based on improved MR-ASM and AAM

A hybrid algorithm of facial feature point location based on...

引用

8th International Conference on Signal Processing, ICSP 2006

作者： Sicong, Zhang Lifang, Wu Xiaoguang, He Jie, Tian School of Electronic Information and Control Engineering Beijing University of Technology Beijing 100022 China Key Lab. of Complex Systems and Intelligence Science Institute of Automation Chinese Academy of Sciences P.O.Box 2728 Beijing 100080 China

ISBN: (纸本)0780397371

Accurate and robust location of feature point is a difficult and complicated issue in face recognition. This paper proposes a facial feature point location algorithm based on improved Multi-Resolution- Active Shape Models (MR-ASM) and Active Appearance Models (AAM). There are three main achievements of our algorithm: 1, we unify MR-ASM and AAM together to improve facial feature point location. 2, AAM is improved so that it has the same implementation framework as ASM. 3, the MR-ASM is adjusted to figure out more robust parameters. Experimental results prove that our algorithm is more accurate than traditional MR-ASM. © 2006 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Energy and thermal aware mapping for mesh-based NoC architectures using multi-objective ant colony algorithm

Energy and thermal aware mapping for mesh-based NoC architec...

引用

IEEE International Conference on Computer Research and Development

作者： Liu, Yanhua Ruan, Ying Lai, Zongsheng Jing, Weiping Dept. of Electronic Computer and Software Systems Institute of Microelectronics Circuit and System East China Normal University Shanghai China Dept. of Electronic Engineering Jiangsu Provincial Key Lab. of ASIC Design Nantong University Nantong China

ISBN: (纸本)9781612848372

Where IP cores to be mapped must be carefully solved for any given application in order to optimize different performance metrics in Network-on-Chip (NoC) design flow. The optimization of different performance metrics simultaneously may cause a negative effect on each other because of the strong correlation between these performance metrics. In this paper, we propose a multi-objective ant colony algorithm (MOACA) that maps IP cores onto mesh-based NoC architectures. This algorithm is an efficient way to find the pareto-optimal front which optimizes energy consumption and hotspot temperature of NoC. The algorithm has been implemented and evaluated for synthetic generated benchmarks. Experimental results confirm the efficiency, accuracy and scalab.lity of the algorithm. © 2011 IEEE.

关键词： Network-on-chip

来源：评论

学校读者我要写书评

暂无评论

A NoSQL based cached storage solution of GIS Web service

International Journal of Digital Content Technology and its ...

引用

International Journal of Digital Content Technology and its Applications 2012年第20期6卷 268-276页

作者： Liu, Jin Wu, Yangping Zhuo, Dan Hu, Qiping Wang, Yan Zhang, Wensheng State Key Lab of Software Engineering Computer School Wuhan University China International School of Software Wuhan University China State Key Lab of Management and Control for Complex Systems Institute of Automation Chinese Academy of Science China

One of challenges in GIS Web service is mass GIS data storage and processing. There are at least three obstacles that oppose dealing with this challenge: the mismatch between the frequent requirement of concurrent tasks for accessing data in GIS Web service and the poor performance of conventional storage mechanism such as relational databases;the insufficiency of relational databases to manage complex data types in GIS Web service;the horizontal scalab.lity of storage for non-stop GIS Web service evolution. This paper proposes a distributed cached storage solution, as the mixture of MongoDB storage and Memcached cache, for GIS Web service. Both techniques are closely related to NoSQL that is a class of database management system identified by its non-adherence to the widely used relational database management system (RDBMS) model. A concise case that facilitates a better understanding of this work indicates its feasibility.

关键词： Websites

来源：评论

学校读者我要写书评

暂无评论

Localization-oriented coverage based on Bayesian estimation in camera sensor networks

Localization-oriented coverage based on Bayesian estimation ...

引用

9th IEEE International Symposium on Wireless, Mobile and Multimedia Networks, WoWMoM 2008

作者： Liu, Liang Zhang, Xi Huadong, Ma. Networking and Information Systems Laboratory Department of Electrical and Computer Engineering Texas A and M University College Station TX 77843 United States Beijing Key Lab. of Intelligent Telecomm Software and Multimedia Beijing University of Posts and Telecommunications Beijing 100876 China

ISBN: (纸本)9781424421008

Target tracking and localization are important applications in camera sensor networks. Although coverage is a very important research topic in wireless sensor networks, and the coverage problem for target detection has been intensively studied, few considers the coverage problem from the perspective of target localization. In this paper, we investigate the coverage problem from the perspective of target localization for camera sensor networks. We first propose a novel localization-oriented sensing model based on the perspective projection of camera. Then, we propose a new notion of Localization-oriented coverage (L-coverage for short). We assume that all camera sensors make the measurements independent of other in the field, and these camera sensors can cooperate to make an accurate estimation for the location of the target. In addition, the relationships among L-coverage, 2-coverage, and the density of camera sensors are also discussed in this paper. The obtained results show that our model can be effectively deployed in many practical scenarios. © 2008 IEEE.

关键词： Target tracking

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：