检索结果-内蒙古大学图书馆

Applied research of data sensing and service to ubiquitous intelligent transportation system

中国高等学校学术文摘·计算机科学 2010年第3期4卷 417-426页

作者： Weifeng LV Bowen DU Dianfu MA Tongyu ZHU Chen WANG State Key Laboratory of Software Development Environment Beihang UniversityBeijing 100191China

High-efficiency transportation systems in urban environments are not only solutions for the growing public travel demands, but are also the premise for enlarging transportation capacity and narrowing the gap between urban and rural areas. Such transportation systems should have characteristics such as mobility, convenience and being accident-free. Ubiquitous-intelligent transportation systems (U-ITS) are next generation of intelligent transportation system (ITS). The key issue of U-ITS is providing better and more efficient services by providing vehicle to vehicle (V2V) or vehicle to infrastructure (V2I) interconnection. The emergence of cyber physical systems (CPS), which focus on information awareness technologies, provides technical assurance for the rapid development of U-ITS. This paper introduces the ongoing Beijing U-ITS project, which utilizes mobile sensors. Realization of universal interconnection between real-time information systems and large-scale detectors allows the system to maximize equipment efficiency and improve transportation efficiency through information services.

关键词： moving sensors (MS) application systems cyber physical systems (CPS) Beijing ubiquitous transportation intelligent system (BUTT)

来源：评论

学校读者我要写书评

暂无评论

Specifying redundancy tactics as crosscutting concerns using aspect-oriented modeling

引用

Frontiers of Computer Science 2014年第6期8卷 977-995页

作者： Xiang QIU Li ZHANG State Key Laboratory of Software Development Environment Beihang University Beijing 100191 China

Various redundancy tactics can be modeled at the design stage of safety-critical systems thereby providing a set of fault-tolerance guidelines for subsequent development activities. However, existing approaches usually interweave redundancy tactics into the functional models making them complex and cluttered; the maintenance of such models is time-consuming and error-prone. To address this problem, we provide a modeling approach to separate the redundancy tactics from the base functional models using aspect-oriented modeling. More specifically, the conceptual models of the redundancy tactics and their semantic constraints are first defined for deriving the relevant aspects. Subsequently, a UML profile is proposed to specify the tactic aspects followed by mapping these concepts to the corresponding concepts of aspect-oriented modeling based on pre-defined principles. In accordance with our proposed profile, reuse directives are applied to handle the overlap of structural features between redundancy tactics and other kinds of tactic. Based on our tactic aspects and their configured attributes, a weaving algorithm is proposed to associate the tactic aspects with the base functional models. The proposed approach is compared with a traditional tactic modeling approach using two safety-critical systems, revealing that： 1） our approach significantly reduces the number of extra model elements needed in the tactic design stage; 2） our approach can largely avoid the impact of changing of the base functional model as the model evolves.

关键词： redundancy tactic, aspect-oriented modeling, weaving algorithm reuse directives

来源：评论

学校读者我要写书评

暂无评论

Improving the Quality of Crowdsourced Image labeling via labelSimilarity

引用

Journal of Computer Science & Technology 2017年第5期32卷 877-889页

作者： Yi-Li Fang Hai-Long Sun Peng-Peng Chen Ting Deng State Key Laboratory of Software Development Environment Beihang University Beijing 100191 China

Crowdsourcing is an effective method to obtain large databases of manually-labeled images, which is especially important for image understanding with supervised machine learning algorithms. However, for several kinds of tasks regarding image labeling, e.g., dog breed recognition, it is hard to achieve high-quality results. Therefore, further optimizing crowdsourcing workflow mainly involves task allocation and result inference. For task allocation, we design a two-round crowdsourcing framework, which contains a smart decision mechanism based on information entropy to determine whether to perform the second round task allocation. Regarding result inference, after quantifying the similarity of all labels, two graphical models are proposed to describe the labeling process and corresponding inference algorithms are designed to further improve the result quality of image labeling. Extensive experiments on real-world tasks in Crowdflower and synthesis datasets were conducted. The experimental results demonstrate the superiority of these methods in comparison with state-of-the-art methods.

关键词： image labeling crowdsourcing information entropy label similarity

来源：评论

学校读者我要写书评

暂无评论

Prune and Merge: Efficient Token Compression for Vision Transformer with Spatial Information Preserved

引用

IEEE Transactions on Multimedia 2025年

作者： Mao, Junzhu Shen, Yang Guo, Jinyang Yao, Yazhou Hua, Xiansheng Shen, Hengtao Nanjing University of Science and Technology School of Computer Science and Engineering Nanjing210094 China Beihang University State Key Laboratory of Software Development Environment Institute of Artificial Intelligence Beijing100191 China Terminus Group Beijing100027 China University of Electronic Science and Technology of China School of Computer Science and Engineering Chengdu611731 China

Token compression is essential for reducing the computational and memory requirements of transformer models, enabling their deployment in resource-constrained environments. In this work, we propose an efficient and hardware-compatible token compression method called Prune and Merge. Our approach integrates token pruning and merging operations within transformer models to achieve layer-wise token compression. By introducing trainable merge and reconstruct matrices and utilizing shortcut connections, we efficiently merge tokens while preserving important information and enabling the restoration of pruned tokens. Additionally, we introduce a novel gradient-weighted attention scoring mechanism that computes token importance scores during the training phase, eliminating the need for separate computations during inference and enhancing compression efficiency. We also leverage gradient information to capture the global impact of tokens and automatically identify optimal compression structures. Extensive experiments on the ImageNet-1k and ADE20K datasets validate the effectiveness of our approach, achieving significant speed-ups with minimal accuracy degradation compared to state-of-the-art methods. For instance, on DeiT-Small, we achieve a 1.64× speed-up with only a 0.2% drop in accuracy on ImageNet-1k. Moreover, by compressing segmenter models and comparing with existing methods, we demonstrate the superior performance of our approach in terms of efficiency and effectiveness. Code and models have been made available at https://***/NUST-Machine-Intelligence-laboratory/prune_and_merge. © 1999-2012 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Tensor decomposition of transportation temporal and spatial big data: A brief review

引用

Fundamental Research 2025年

作者： Li, Linchao Lin, Xiang Ran, Bin Du, Bowen National Key Laboratory of Green and Long-Life Road Engineering in Extreme Environment (Shenzhen) Shenzhen University Shenzhen 518060 China Department of Civil and Environmental Engineering University of Wisconsin 53706 Madison United States State Key Laboratory of Software Development Environment Beihang University Beijing 100191 China Key Laboratory of Structural Health Monitoring and Control Shijiazhuang Tiedao University Shijiazhuang 050043 China

Recent development in sensing and communication technologies has made the collection of a large amount of traffic data easy and transportation engineering has entered the big data era. The massive traffic data provides some good opportunities for Intelligent Transportation System (ITS), while some great challenges because of its characteristics of large value, variety, velocity, veracity, and volume. In recent few years, tensor decomposition has played an important role in traffic data analytic solutions and attached great interest from both academic and industrial areas. In this paper, the preliminary background and the implementation of tensor decomposition are presented. Then, some recent studies of tensor decomposition for traffic data imputation, traffic state prediction, and analysis of travel pattern are reviewed. Furthermore, advantages and disadvantages are discussed. Finally, remaining challenges of the application of tensor decomposition in transportation engineering are pointed out. © 2024

关键词： Data mining Intelligent transportation system Survey Traffic prediction

来源：评论

学校读者我要写书评

暂无评论

REPRESENTATION AND AUTOMATED TRANSFORMATION OF GEOMETRIC stateMENTS

引用

Journal of Systems Science & Complexity 2014年第2期27卷 382-412页

作者： CHEN Xiaoyu State Key Laboratory of Software Development Environment School of Computer Science and EngineeringBeihang University

The emergence of a large quantity of digital resources in geometry, various geometric automated theorem proving systems, and kinds of dynamic geometry software systems has made geometric computation, reasoning, drawing, and knowledge management dynamic, automatic or interactive on computer. Integration of electronic contents and different systems is desired to enhance their accessibility and exploitability. This paper proposes an equivalent transformation framework for manipulating geometric statements available in the literature by using geometry software systems. Such a framework works based on a newly designed geometry description language(GDL), in which geometric statements can be represented naturally and easily. The author discusses and presents key procedures of automatically transforming GDL statements into target system-native representations for *** author also demonstrates the framework by illustrating equivalent transformation processes and interfaces for compiling the transformation results into executable formats that can be interpreted by the target geometry software systems for automated theorem proving and dynamic diagram drawing.

关键词： Automated diagram drawing equivalent transformation formalized geometric statements geometric automated theorem proving.

来源：评论

学校读者我要写书评

暂无评论

Accelerating MUS enumeration by inconsistency graph partitioning

引用

Science China(Information Sciences) 2019年第11期62卷 70-80页

作者： Jie LUO Shaofan LIU State Key Laboratory of Software Development Environment School of Computer Science and EngineeringBeihang University

The problem of finding minimal unsatisfiable subsets(MUSes) has been studied frequently because of its theoretical importance and wide range of applications in domains such as electronic design automation, software, and integrated circuit verification. In this paper, a method for accelerating the enumeration of MUSes based on inconsistency graph partitioning is proposed. First, an inconsistency graph of a set of clauses is constructed by extracting the inconsistency relations between literals of different clauses. In this paper, we show that by partitioning the inconsistency graph into small connected components through a vertex cut, the enumeration of MUSes in different components becomes independent and it is possible to compute them separately. Moreover, the MUSes of the original clause set can be constructed by merging the unit clauses in the MUSes of these connected components back into the clauses in the vertex cut. Experiments show that by integrating the acceleration method into the MARCO MUSes enumerator, there is a 2–3 times improvement in the average runtime of solved instances for randomly generated benchmarks. By integrating the acceleration method into itself as an MUS enumerator, there is another 3–4 times improvement when compared with the accelerated MARCO.

关键词： minimal unsatisfiable subsets inconsistency graph graph partition SAT UNSAT

来源：评论

学校读者我要写书评

暂无评论

Recommender systems based on ranking performance optimization

引用

Frontiers of Computer Science 2016年第2期10卷 270-280页

作者： Richong ZHANG Han BAO Hailong SUN Yanghao WANG Xudong LIU State Key Laboratory of Software Development Environment Beihang University B eijing 100191 China

The rapid development of online services and information overload has inspired the fast development of recommender systems, among which collaborative filtering algorithms and model-based recommendation approaches are wildly exploited. For instance, matrix factorization （MF） demonstrated successful achievements and advantages in assisting internet users in finding interested information. These existing models focus on the prediction of the users＇ ratings on unknown items. The performance is usually evaluated by the metric root mean square error （RMSE）. However, achieving good performance in terms of RMSE does not always guarantee a good ranking performance. Therefore, in this paper, we advocate to treat the recommendation as a ranking problem. Normalized discounted cumulative gain （NDCG） is chosen as the optimization target when evaluating the ranking accuracy. Specifically, we present three ranking-oriented recommender algorithms, NSME AdaMF and AdaNSME NSMF builds a NDCG approximated loss function for Matrix Factorization. AdaMF is based on an algorithm by adaptively combining component MF recommenders with boosting method. To combine the advantages of both algorithms, we propose AdaNSME which is a hybird of NSMF and AdaME and show the superiority in both ranking accuracy and model generalization. In addition, we compare our proposed approaches with the state-of-the-art recommendation algorithms. The comparison studies confirm the advantage of our proposed approaches.

关键词： recommender system matrix factorization,learning to rank

来源：评论

学校读者我要写书评

暂无评论

Personalized Tag Recommendation Using Social Influence

引用

Journal of Computer Science & Technology 2012年第3期27卷 527-540页

作者：胡军王兵刘禹李德毅 State Key Laboratory of Software Development Environment Beihang University

Tag recommendation encourages users to add more tags in bridging the semantic gap between human concept and the features of media object, which provides a feasible solution for content-based multimedia information retrieval. In this paper, we study personalized tag recommendation in a popular online photo sharing site Flickr. Social relationship information of users is collected to generate an online social network. From the perspective of network topology, we propose node topological potential to characterize user＇s social influence. With this metric, we distinguish different social relations between users and find out those who really have influence on the target users. Tag recommendations are based on tagging history and the latent personalized preference learned from those who have most influence in user＇s social network. We evaluate our method on large scale real-world data. The experimental results demonstrate that our method can outperform the non-personalized global co-occurrence method and other two state-of-the-art personalized approaches using social networks. We also analyze the further usage of our approach for the cold-start problem of tag recommendation.

关键词： recommendation system social tagging personalization social network Flickr

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Optimization Algorithm for the CMST Problem

引用

电子学报(英文版) 2011年第4期20卷 583-589页

作者： HAN Jun FANG Qingan MAO Liyong HUANG Yaling State Key Laboratory of Software Development Environment Beihang University Beijing China

The Capacitated minimum spanning tree (CMST) problem is one of the most fundamental and significant problems in the optimal design of networks. It is also a classical combinatorial optimization problem which has been tackled by researchers for centuries using various methods. In this paper, a new NS-TS hybrid optimization algorithm that combines neighborhood search and tabu search is proposed. A novel neighborhood structure and associate tabu strategy is proposed and implemented. Computational experiments showing the effectiveness and efficiency of the algorithm on benchmark instances are given.

关键词：混合优化算法组合优化问题禁忌搜索最小生成树优化设计研究人员网络

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：