检索结果-内蒙古大学图书馆

2009 2nd International Congress on Image and Signal Processing, CISP'09

作者： Zhongli, Wang Xueshan, Gao Baochang, Zhang Hao, Wu State Key Lab. of Robotics and System Harbin Institute of Technology Harbin China School of Computer Science Beijing Institute of Technology Beijing China Intelligent Robotics Institute Beijing Institute of Technology Beijing China School of Automation Science and Electric Engineering Beihang University Beijing China

ISBN: (纸本)9781424441310

Vision-based motion estimation is a key technique in autonomously mobile robot. This paper presents an integrated method for motion estimation of the mobile robot based on floor plane extraction. The whole processes include floor plane extraction, pose estimation of the onboard stereo vision system and motion estimation of the mobile robot. First, based on disparity map obtained from the onboard real-time stereo vision system, a two-stage method is used for fast floor plane extraction. Then all the points belong to the floor plane are transformed from disparity space to Euclidean space, and a simple least squares method is employed to fit the floor plane. The fitting parameters of the floor plane are used for the pose estimation of the onboard stereo vision system. Finally, the motion of the mobile robot is estimated with the tracking results of the points on floor plane. The proposed motion estimation method is intended to be used in the motion control of a wheeled mobile robot. Experimental results in real environments are presented. ©2009 IEEE.

关键词： Mobile robots

来源：评论

学校读者我要写书评

暂无评论

Attention-Based Knowledge-aware Multi-Interest Intelligent Model for Sequential Recommendation 19

Attention-Based Knowledge-aware Multi-Interest Intelligent M...

引用

2022 IEEE SmartWorld, 19th IEEE International Conference on Ubiquitous Intelligence and Computing, 2022 IEEE International Conference on Autonomous and Trusted Vehicles Conference, 22nd IEEE International Conference on Scalab.e Computing and Communications, 2022 IEEE International Conference on Digital Twin, 8th IEEE International Conference on Privacy Computing and 2022 IEEE International Conference on Metaverse, SmartWorld/UIC/ATC/ScalCom/DigitalTwin/PriComp/Metaverse 2022

作者： Li, Yang Li, Qianmu Meng, Shunmei Hou, Jun Nanjing University of Science and Technology School of Computer Science and Engineering Nanjing China Wuyi University Intelligent Manufacturing Department Jiangmen China Nanjing University State Key Lab. for Novel Software Technology Nanjing China Nanjing Vocational University of Industry Technology School of Social Science Nanjing China

ISBN: (纸本)9798350346558

As a fundamental means of intelligent service, recommender systems have shown a trend of diversification. Sequential recommendation, a class of time-aware methods, aims to predict the user's next click. Multi-interest representation is a hot topic in recommender systems, but the research on how to apply multi-interest into sequences is still in the minority. And many methods will introduce knowledge graph to improve the representation of items, but how to consider both knowledge graph and user's multi-interest in sequential recommendation has become a temporary blank. In this paper, we propose an attentive sequential recommendation with knowledge-aware multi-interest module (AKAMI). Items are embedded from the sequence's structure and user's interests. Specifically, we represent items by stacking self-attention layers to mine the internal correlation of the sequence. Meanwhile, we segment the sequence via the common neighbors in the knowledge graph and cast the user's multi-interest based on the aggregation of subsequences. Multi-interests are integrated into item representation by the gating network adaptively. Comprehensive experiments on two real-world datasets show that the proposed model performs better than various representative methods under different evaluation metrics. Additionally, we also visually explain the model through a case study. © 2022 IEEE.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Concrete Structural Crack Damage Classification Using Nonlinear Dimension Reduction and Broad Learning system

Concrete Structural Crack Damage Classification Using Nonlin...

引用

2024 IEEE International Conference on systems, Man, and Cybernetics, SMC 2024

作者： Wang, Bingshu Lin, Jia Zhuang, Xiaodong Zhang, Guanghui Chen, C.L. Philip School of Software Northwestern Polytechnical University Xi'an China Shenzhen University Guangdong Key Lab. of Intelligent Info. Processing and Shenzhen Key Laboratory of Media Security Shenzhen China School of Computer Science and Technology Shandong University Qingdao China School of Computer Science and Engineering South China University of Technology Guangzhou China

ISBN: (纸本)9781665410205

Concrete structural crack damage classification is of importance for road safety. This paper proposes a new method based on broad neural network for crack damage classification in concrete structures. It includes three stages. Firstly, a pre-trained deep neural network is used to extract the features from crack images. Secondly, principal component analysis is used to project the retrieved features from high dimensions to low dimensions. Thirdly, broad learning system is employed to predict the classification using the low-dimensional features. Experimental results demonstrate that this method reduces the model's training time and improves classification accuracy. © 2024 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Combing data filter and data sampling for cross- Company defect prediction: An empricial study 29

Combing data filter and data sampling for cross- Company def...

引用

29th International Conference on software Engineering and Knowledge Engineering, SEKE 2017

作者： Yu, Xiao Wu, Man Zhang, Yan Fu, Mandi State Key Lab. of Software Engineering Computer School Wuhan University Wuhan China School of Computer Science and Information Engineering HuBei University Wuhan China Educational Informationalization Engineering Research Center of HuBei Province Wuhan China Wuhan National Laboratory for Optoelectronics Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)1891706411

Cross-company defect prediction (CCDP) is a practical way that trains a prediction model by exploiting one or multiple projects of a source company and then applies the model to target company. Unfortunately, larger irrelevant crosscompany (CC) data usually makes it difficult to build a prediction model with high performance. On the other hand, the CC data has the highly imbalanced nature between the defectiveprone and non-defective classes, which will degrade the performance of CCDP. To address such issues, this paper proposes an approach, in which data sampling is combined with data filter, to overcome these problems. Data sampling seeks a more balanced dataset through the addition or removal of instances, while data filter is a process of filtering out the irrelevant CC data so that the performance of CCDP models can be improved. We employ two data filtering methods called NN filter and DBSCAN filter combined with SMOTE (Synthetic Minority Oversampling Technique) and RUS (Random Under- Sampling). Eight different approaches would be produced when combing these four techniques: 1- NN filter performed prior to RUS;2- NN filter performed after RUS;3- NN filter performed prior to SMOTE;4- NN filter performed after SMOTE;5- DBSCAN filter performed prior to RUS;6- DBSCAN filter performed after RUS;7- DBSCAN filter performed prior to SMOTE;8- DBSCAN filter performed after SMOTE. The empirical study was carried out on 15 publicly availab.e project datasets. The experimental results demonstrate that NN filter performed prior to RUS (Approach 1) performs better than the other seven approaches.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Quantified Dynamic Metric Temporal Logic for Dynamic Networks of Stochastic Hybrid Automata

Quantified Dynamic Metric Temporal Logic for Dynamic Network...

引用

International Conference on Application of Concurrency to system Design

作者： Alexandre David Kim G. Larsen Axel Legay Guangyuan Li Danny Bøgsted Poulsen Department of Computer Science Aalborg University State Key Lab. of Computer Science Chinese Academy of Sciences

Multiprocessing systems are capable of running multiple processes concurrently. By now such systems have established themselves as the defacto standard for operating systems. At the core of an operating system is the ability to execute programs and as such there must be a primitive for instantiating new processes - also programs are allowed to die/terminate. Operating systems may allow the executing programs to split (spawn) into more computational threads in order to let programs take advantage of concurrent execution as well. One of the most used modelling languages, Timed Automata, is based on multiple automata interacting thus they easily model the concurrent execution of programs. However, this language assumes a fixed size system in the sense that automata cannot be created at will but must be instantiated when the overall system is created. This is in contrast with the fact that developers are able to create threads when needed. In this paper we present our continued work to incorporate spawning of threads into UPPAAL SMC. Our modelling language, Dynamic Networks of Stochastic Hybrid Automata, is essentially Timed Automata extended with a spawning primitive and a tear-down primitive. The dynamic creation of threads has the side-effect that it is no longer possible to use ordinary logics to specify behaviours of individual threads - because the threads no longer have unique names. In this paper we propose an extension of Metric Temporal Logic with means for quantifying over the dynamically created threads. This makes it possible to zoom in on individual threads and specify requirements to their future behaviour. Furthermore, we present a monitoring procedure for the logic based on rewriting formulas. The presented modelling language and the specification language have been implemented in UPPAAL SMC version 4.1.18.

关键词： Servers Automata Instruction sets Semantics Ports (computers) Delays Stochastic processes

来源：评论

学校读者我要写书评

暂无评论

Detecting infeasible traces in process models

Detecting infeasible traces in process models

引用

14th International Conference on Enterprise Information systems, ICEIS 2012

作者： Wang, Zhaoxia Wen, Lijie Zhu, Xiaochen Liu, Yingbo Wang, Jianmin School of Software Tsinghua University Beijing 100084 China Department of Computer Science and Technology Tsinghua University Beijing 100084 China Key Lab. for Information System Security Ministry of Education Beijing 100084 China National Laboratory for Information Science and Technology Beijing 100084 China Logistical Engineering University Chongqing 400016 China

ISBN: (纸本)9789898565105

Workflow testing is an important method of workflow analysis in design time. A challenging problem with trace-oriented test data generation in particular and trace-based workflow analysis in general is the existence of infeasible traces for which there is no input data for them to be executed. In this paper we build on the theory of workflow nets and introduce workflow nets where transitions have conditions associated with them. We then demonstrate that we can determine which execution traces, that are possible according to the control-flow dependencies, are actually possible taking the data perspective into account. This way we are able to more accurately determine in design time the infeasible traces caused by the correlation between transition conditions along this trace. Finally, we provide a solution to automatically detecting the shortest infeasible trace.

关键词： Infeasible traces Process models Workflow analysis

来源：评论

学校读者我要写书评

暂无评论

Using class imbalance learning for cross-company defect prediction 29

Using class imbalance learning for cross-company defect pred...

引用

29th International Conference on software Engineering and Knowledge Engineering, SEKE 2017

作者： Yu, Xiao Zhou, Mingsong Chen, Xu Deng, Lijun Wang, Lu State Key Lab. of Software Engineering Computer School Wuhan University Wuhan China Department of Information Science and Technology Beijing Normal University Beijing China College of Business City University of Hong Kong Kowloon Tong Hong Kong School of Computer Science and Information Engineering HuBei University Wuhan China

ISBN: (纸本)1891706411

Cross-company defect prediction (CCDP) is a practical way that trains a prediction model by exploiting one or multiple projects of a source company and then applies the model to target company. Unfortunately, the performance of such CCDP models is susceptible to the high imbalanced nature between the defect-prone and non-defect classes of CC data. Class imbalance learning is applied to alleviate this issue. Because many class imbalance learning methods have been proposed, there is an imperative need to analyze and compare the performance of these methods for CCDP. Although prior empirical studies have proven *** algorithm achieves the best performance for defect prediction. This observation leads us to conduct a careful empirical study the issues of if and how class imbalance learning methods can benefit cross-company defect prediction. We investigate different types of class imbalance learning methods, including under-sampling technique, over-sampling technique and over sampling followed by under-sampling technique on the cross-company defect prediction performance over 15 publicly availab.e datasets. Experimental results show that under-sampling technique achieves the best overall performance in terms of the gmeasure among those methods we studied.1.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Automatic semantic web service composition via agent intention execution in AgentSpeak 07

Automatic semantic web service composition via agent intenti...

引用

IEEE/WIC/ACM International Conference on Web Intelligence, WI 2007

作者： Li, Huan Yu, Fan Qin, Zheng Qin, Jun Yang, Bo Dep. of Computer Science and Technology Xian JiaoTong University Dongguan University of Technology Key Lab. for ISS of MOE Software School Tsinghua University Dep. of Computer Science and Technology Xian JiaoTong University Tsinghua University Manchester Business School Dep. of Computer Science and Technology Xian JiaoTong University

ISBN: (纸本)0769530265

AI planning is the main stream method for automatic semantic web service composition (SWSC) research. However, planning based SWSC method can only return service composition upon user requirement description and lacks flexibility to deal with environment change. Deliberate agent architecture, such as BDI agent, is hopeful to make SWSC more intelligent. In this paper, we propose an automatic SWSC enabling method for AgentSpeak agent. Firstly, conversion algorithm from OWL-S web service description to agent's plan set (OWLS2APS) is presented. Target service is converted to agent's goal and related services are converted into agent's plan set. Then, SWSC is automatically performed through agent's intention formation. Agent invokes web service according to service sequence converted back from its intention. Agent can behave rationally with rules or ask for human intervention when SWSC or service invocation is not feasible. At last, a case study on enterprise credit rating service composition is presented to illustrate the method. © 2007 IEEE.

关键词： Web services

来源：评论

学校读者我要写书评

暂无评论

Dynamic Detection of AsyncTask Related Defects

Dynamic Detection of AsyncTask Related Defects

引用

IEEE International Conference on software Quality, Reliability and Security (QRS)

作者： Qing Liu Linjie Pan Baoquan Cui Jun Yan Jian Zhang State Key Lab. of Computer Science Institute of Software CAS University of Chinese Academy of Sciences Technology Center of Software Engineering Institute of Software CAS

ISBN: (纸本)9781665458146

As a widely used Android asynchronous component, AsyncTask is used to run time-consuming tasks. However, the misuse of AsyncTask will cause defects, i.e., crashes and memory leaks. Based on static analysis, existing approaches cannot accurately detect AsyncTask-related defects and produce many false positives since some paths are not reachable in practice. In this paper, we propose a dynamic detection method based on instrumentation, Monkey execution and log analysis to detect these defects. And we implement a tool AD2Checker based on the proposed method. Our experiment on 19 real-world apps shows that it has found 145 bugs and has no false positives. Moreover, it triggers crashes caused by misuse of AsyncTask.

关键词： Printing Instruments Conferences computer bugs Static analysis software quality software reliability

来源：评论

学校读者我要写书评

暂无评论

RFID Reader Planning for the Surveillance of Predictable Mobile Objects

引用

Procedia computer science 2018年 129卷 475-481页

作者： Weiping Zhu Mingzhe Li School of Computer Science Wuhan University P.R. China State Key Lab. for Novel Software Technology Nanjing University P.R. China

The surveillance of mobile objects is important for many applications in public security, logistics, traffic control, etc. Multiple RFID (Radio frequency identification) readers can be placed in the given region to fulfill this work. However, due to the mobility of objects and various kinds of RFID collisions, a proper planning of RFID readers is needed. Existing RFID reader planning approaches are designed for stationary scenarios and hence cannot have desirable performance when handling mobile objects. In this paper, we investigate the RFID reader planning problem for the surveillance of mobile objects whose motion can be predicted. We define the evaluation metric for the reader planning by dividing the surveillance time into multiple time slots and compute the successful identification of tags in individual time slot respectively. We further propose several reader planning strategies to handle RFID collisions in the surveillance. Genetic algorithm and particle swarm optimization are used to get the optimal result based on the evaluation metric and the reader planning strategies. We perform extensive simulations for validating the proposed approaches. The results show that our approach can achieve a good accuracy in the surveillance.

关键词： RFID Surveillance Reader Planning Mobile Objects Predictable

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：