检索结果-内蒙古大学图书馆

27th International Conference on Pattern Recognition, ICPR 2024

作者： Xie, Zaipeng Zhang, Xiangqin Wang, Yunfei Jie, Xuanyao Fang, Wenhao Cai, Yanping Key Laboratory of Water Big Data Technology of Ministry of Water Resources Hohai University Nanjing China College of Computer Science and Software Engineering Hohai University Nanjing China

ISBN: (纸本)9783031781827

Runoff forecasting plays a crucial role in water resource management and flood mitigation, but it often faces significant challenges due to data deficiency and decentralized datasets. Inadequate hydrological data in many watersheds hinders the development of accurate prediction models, while organizational barriers and concerns about data privacy lead to information silos, preventing effective collaborative modeling. To address these issues, we propose FedHydroDSW, a novel personalized federated learning framework for runoff forecasting in data-scarce watersheds. Our approach enables decentralized model training across multiple organizations without requiring direct data sharing, thus preserving data privacy. FedHydroDSW incorporates unique model similarity metrics and parameter adaptation methods, utilizing insights from data-rich areas to refine forecasts in data-scarce regions. Experiments on the CAMELS dataset show that FedHydroDSW significantly outperforms standalone and conventional federated learning methods, with an average 30.1% increase in NSE, and reductions of 28% in RAE and 26.6% in RMSE across multiple data-scarce basins. By enabling secure cross-silo learning, our FedHydroDSW strategy paves the way toward more equitable runoff forecasting globally. The customized training and pattern recognition techniques represent major progress in deploying federated intelligence for impactful hydrological predictions. Our code is publicly available at https://***/xyjie37/FedHydroDSW. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

EventLens: Enhancing Visual Commonsense Reasoning by Leveraging Event-Aware Pretraining and Cross-modal Linking

EventLens: Enhancing Visual Commonsense Reasoning by Leverag...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Ma, Mingjie Yu, Zhihuan Ma, Yichao Li, Guohui Yang, Zhong School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China School of Software Engineering Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)9798350368741

Visual Commonsense Reasoning (VCR) is a cognitive task, challenging models to answer visual questions, and to explain the rationale behind their answers. While Large Language Models (LLMs) offer potential for this task, VCR's complex scenes require specialized approaches to activate their commonsense reasoning abilities, as existing Multimodal LLMs struggle with VCR's visual events and unique reference tags. To address these challenges, we propose EventLens, which enhances VCR through Event-Aware Pretraining and Cross-modal Linking in Supervised Fine-tuning. First, we introduce a new pretraining stage that emulates human cognitive processes to improve LLM comprehension of complex scenarios. Second, during supervised fine-tuning, we leverage reference tags to explicitly bridge RoI features with text, maintaining semantic integrity across modalities. Additionally, instruct prompts and task-specific adapters help integrate LLMs' knowledge with new commonsense reasoning. Experimental results demonstrate competitive performance with state-of-the-art methods and ablation studies verify the effectiveness of proposed EventLens. © 2025 IEEE.

关键词： Multimedia Understanding Pretrained Language Models Visual Commonsense Reasoning

来源：评论

学校读者我要写书评

暂无评论

Understanding and Identifying Technical Debt in the Co-Evolution of Production and Test Code

引用

IEEE Transactions on software engineering 2025年第05期51卷 1415-1436页

作者： Guo, Yimeng Chen, Zhifei Xiao, Lu Chen, Lin Li, Yanhui Zhou, Yuming Nanjing University State Key Laboratory for Novel Software Technology Nanjing210023 China Nanjing University of Science and Technology School of Computer Science and Engineering Nanjing210094 China Stevens Institute of Technology School of Systems and Enterprises HobokenNJ07030 United States

The co-evolution of production and test code (PT co-evolution) has received increasing attention in recent years. However, we found that existing work did not comprehensively study various PT co-evolution scenarios, such as the qualification and persistence of their effects on software. Inspired by technical debt (TD), we refer to TD generated during the co-evolution between production and test code as PT co-evolution technical debt (PTCoTD). To better understand PT co-evolution, we first conducted an exploratory study on its characteristics on 15 open-source projects, finding unbalanced PT co-evolution prevalent and summarizing five potential PT flaws. Then we proposed an approach to identify and quantify PTCoTDs of these flaw patterns, considering evolutionary and structural relationships. We also built prediction models to describe cost trajectories and rank all PTCoTDs to prioritize expensive ones. The evaluation on the 15 projects shows that our approach can identify PTCoTDs that deserve attention. The identified PTCoTDs account for about half of the project's total maintenance costs, and the cost proportion of the expensive Top-5 is 1.8x more than the file proportion they contain. Almost all covered maintenance costs persist as PTCoTD in the future, with an average increase of 6.8% between the last two releases. Our approach also accurately predicts the costs of PTCoTD with an average prediction deviation of only 8.3%. Our study provides valuable insights into PT co-evolution scenarios and their effects, which can guide practices and inspire future work on software testing and maintenance. © 2025 IEEE.

关键词： software testing

来源：评论

学校读者我要写书评

暂无评论

Task Consistent Prototype Learning for Incremental Few-Shot Semantic Segmentation 27th

Task Consistent Prototype Learning for Incremental Few-Shot...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Xu, Wenbo Wu, Yanan Jiang, Haoran Wang, Yang Wu, Qiang Zhang, Jian Faculty of Engineering and Information Technology University of Technology Sydney Sydney2007 Australia School of Computer and Information Technology Beijing Jiaotong University Beijing100044 China Department of Computer Science and Software Engineering Concordia University MontrealH3G2J1 Canada

ISBN: (纸本)9783031783463

Incremental Few-Shot Semantic Segmentation (iFSS) tackles a task that requires a model to continually expand its segmentation capability on novel classes using only a few annotated examples. Typical incremental approaches encounter a challenge that the objective of the base training phase (fitting base classes with sufficient instances) does not align with the incremental learning phase (rapidly adapting to new classes with less forgetting). This disconnect can result in suboptimal performance in the incremental setting. This study introduces a meta-learning-based prototype approach that encourages the model to learn how to adapt quickly while preserving previous knowledge. Concretely, we mimic the incremental evaluation protocol during the base training session by sampling a sequence of pseudo-incremental tasks. Each task in the simulated sequence is trained using a meta-objective to enable rapid adaptation without forgetting. To enhance discrimination among class prototypes, we introduce prototype space redistribution learning, which dynamically updates class prototypes to establish optimal inter-prototype boundaries within the prototype space. Extensive experiments on iFSS datasets built upon PASCAL and COCO benchmarks show the advanced performance of the proposed approach, offering valuable insights for addressing iFSS challenges. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Implementation of Lean Management in software Project Management

Implementation of Lean Management in Software Project Manage...

引用

International Conference on Advancements in Computational Sciences (ICACS)

作者： Uzair Rasheed Sadia Khalid Mishal Muneer Israr Ahmad Ibraheem Khalil Faisal Masood Department of Computer and Software Engineering College of Electrical and Mechanical Engineering NUST Islamabad Pakistan

ISBN: (数字)9798331533038

ISBN: (纸本)9798331533045

Organizations normally focus on delivering the product that could earn maximum profit for the client. They do not focus on time, cost and quality. So the resultant product is of poor quality, over budget, over time, poorly satisfies the client’s requirements, and is high in cost. In this research paper, we will use lean concepts in software project management. It will not only remove all the waste in project management but also improve the project’s quality, time, and cost. We will also compare the project management concept with the lean project management concept and how combining these two will give us a perfect idea to increase software quality, reduce time, and deliver faster delivery. The only drawback is that current standards of software project management are not mature enough to adopt lean management to its extent.

关键词： Waste management Costs Standards organizations Decision making Project management software quality Organizations software reliability software development management

来源：评论

学校读者我要写书评

暂无评论

KAN-HyperpointNet for Point Cloud Sequence-Based 3D Human Action Recognition

KAN-HyperpointNet for Point Cloud Sequence-Based 3D Human Ac...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Chen, Zhaoyu Li, Xing Huang, Qian Geng, Qiang Yang, Tianjin Han, Shihao College of Information Science and Technology & Artificial Intelligence Nanjing Forestry University China College of Computer Science and Software Engineering Hohai University Nanjing China

ISBN: (纸本)9798350368741

Point cloud sequence-based 3D action recognition has achieved impressive performance and efficiency. However, existing point cloud sequence modeling methods cannot adequately balance the precision of limb micro-movements with the integrity of posture macro-structure, leading to the loss of crucial information cues in action inference. To overcome this limitation, we introduce D-Hyperpoint, a novel data type generated through a D-Hyperpoint Embedding module. D-Hyperpoint encapsulates both regional-momentary motion and global-static posture, effectively summarizing the unit human action at each moment. In addition, we present a D-Hyperpoint KANsMixer module, which is recursively applied to nested groupings of D-Hyperpoints to learn the action discrimination information and creatively integrates Kolmogorov-Arnold Networks (KAN) to enhance spatio-temporal interaction within D-Hyperpoints. Finally, we propose KAN-HyperpointNet, a spatio-temporal decoupled network architecture for 3D action recognition. Extensive experiments on two public datasets: MSR Action3D and NTU-RGB+D 60, demonstrate the state-of-the-art performance of our method. © 2025 IEEE.

关键词： 3D action recognition D-Hyperpoint KAN-HyperpointNet point cloud sequence

来源：评论

学校读者我要写书评

暂无评论

Automated software Requirements Prioritization Using Natural Language Processing

Automated Software Requirements Prioritization Using Natural...

引用

International Conference on Advancements in Computational Sciences (ICACS)

作者： Israr Ahmad Uzair Rasheed Ibrahim Khalil Zaeem Anwaar Wasi Haider Butt Department of Computer and Software Engineering College of Electrical and Mechanical Engineering NUST Islamabad Pakistan

ISBN: (数字)9798331533038

ISBN: (纸本)9798331533045

The software requirements specification (SRS) can become a problem & barrier to the successful completion of a project, this can become an obstacle to the successful delivery of a project through the software requirements specification (SRS). In some cases, this results in not addressing real needs. The SRS dataset can include duplicate data or disputed content which cause high costs and wastage of the time and the general effectiveness of the project will be reduced. Latest advancements in machine learning have led to active endeavors to formulate automated approaches for generating seamless software requirements specification [6]. We thus apply transformer models, such as BERT and RobERTa in this study, for classification. We examine multiclass text classification with RoBERTa and classification tasks involving the prediction of type, priority, and severity of user-specified requirements. Along with that, we compare its performance to other natural language processing (NLP) models like LSTM and BiLSTM. Our experiments on the DOORS dataset demonstrate higher accuracy using RoBERTa compared to existing approaches. Assessment parameters included accuracy, F1 score, recall, and precision.

关键词： Deep learning Accuracy Costs Computational modeling Text categorization Bidirectional long short term memory Transformers software Natural language processing

来源：评论

学校读者我要写书评

暂无评论

A Credible and Fair Federated Learning Framework Based on Blockchain

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年第2期6卷 301-316页

作者： Chen, Leiming Zhao, Dehai Tao, Liping Wang, Kai Qiao, Sibo Zeng, Xingjie Tan, Chee Wei China University of Petroleum School of Computer Science and Technology East China Qingdao266580 China CSIRO's Data61 Sydney2015 Australia Hefei University of Technology Anhui230009 China Nanyang Technological University School of Computer Science and Engineering Singapore639798 Singapore Tiangong University School of Software Tianjin300387 China Southwest Petroleum University School of Computer Science Chengdu610500 China

Federated learning (FL) enables cooperative computation between multiple participants while protecting user privacy. Currently, FL algorithms assume that all participants are trustworthy and their systems are secure. However, the following problems arise in real-world scenarios: 1) Malicious clients disrupt FL through model poisoning and data poisoning attacks. Although some research has proposed secure aggregation methods to solve this problem, most methods have limitations. 2) The current method cannot fairly evaluate client contribution in some scenarios. Some clients exhibit free-rider behavior, seeking to cheat the reward system and manipulate global models. Evaluating client contribution and distributing rewards also present challenges. To address these challenges, we design a trustworthy federated framework to ensure secure computing throughout the federated task process. First, we propose a method of detecting malicious models to guarantee secure model aggregation. Then, we propose a fair method of assessing contribution to identify client-side free-riding behavior. Finally, we implement a computation process based on blockchain and smart contracts to guarantee the trustworthiness and fairness of federated tasks. To validate the performance of our framework, we simulate different types of client attacks and contribution evaluation scenarios on several open-source datasets. The experiments show that our framework ensures the credibility of federated tasks and achieves a fair evaluation of client contributions. © 2020 IEEE.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Deeper Anomaly Detection in Industrial Automation

Deeper Anomaly Detection in Industrial Automation

引用

IEEE Annual Computing and Communication Workshop and Conference (CCWC)

作者： Ho-Seok Jung Sung Won Jang Youngmi Baek Dept. of Computer Software Engineering Changshin University Korea Dept. of Computer Engineering Changshin University Korea

ISBN: (数字)9798331507695

ISBN: (纸本)9798331507701

The security of industrial networks, particularly in industrial automation systems, is critical for ensuring system reliability and protecting sensitive data. This paper proposes a deeper anomaly detection system using the ResNet34 (Residual Network) model to identify and detect cyber-attacks in industrial networks, specifically focusing on Controller Area Network (CAN) systems. The study highlights the vulnerabilities in industrial communication protocols, such as CAN, Modbus, and Ethernet/IP, which are susceptible to cyber-attacks including replay, modification, and fuzzing attacks. These attacks can disrupt the functioning of industrial systems and cause significant damage. Experimental results show that the proposed model achieves a 100 % detection rate for all types of cyber-attacks, demonstrating its effectiveness in recognizing abnormal patterns and responding to changes in network behavior. The results confirm that the ResNet34-based deep anomaly detection model can be a valuable tool for strengthening the security of industrial networks by providing real-time detection of cyber-attacks, thereby ensuring the stability and safety of industrial automation systems.

关键词： Automation Stability analysis Real-time systems Safety Controller area networks Security Reliability Cyberattack Anomaly detection Residual neural networks

来源：评论

学校读者我要写书评

暂无评论

A Review on Feature Selection Techniques for Anxiety Classification from Physiological Signals 1

A Review on Feature Selection Techniques for Anxiety Classif...

引用

1st International Conference on Advances in computer Science, Electrical, Electronics, and Communication Technologies, CE2CT 2025

作者： Shikha, Shikha Sethia, Divyashikha Indu, S. Delhi Technological University Department of Computer Science and Engineering New Delhi India Delhi Technological University Department of Software Engineering New Delhi India Delhi Technological University Department of Electronics and Communication New Delhi India

ISBN: (纸本)9798331518578

Anxiety disorders significantly impact individuals' quality of life and are traditionally assessed using subjective methods, which often lack accuracy. Recent advancements have enabled the use of physiological signals for anxiety classification, providing a more objective approach. However, the multidomain nature of these signals results in high-dimensional feature sets, necessitating effective feature selection. Therefore, this paper reviews various feature selection techniques-filter, wrapper, embedded, and hybrid-used in physiological-based anxiety classification, analyzing the advantages and limitations. It also highlights key challenges and suggests future research directions. Furthermore, this review introduces a proposed study utilizing a novel Corr-FGA hybrid feature selection algorithm to develop an efficient anxiety classification framework. The proposed framework achieves the highest accuracy of 88.93% using 8 features. This review aims to guide researchers in developing a robust, efficient anxiety classification model while providing insights that can help streamline feature selection and optimize machine learning model performance in real-world applications. © 2025 IEEE.

关键词： Adversarial machine 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：