检索结果-内蒙古大学图书馆

Edge Guided Network with Motion Enhancement for Few-Shot Action Recognition

IEEE Transactions on Circuits and Systems for Video Technology 2025年第6期35卷 5331-5342页

作者： Du, Kaiwen Ye, Weirong Guo, Hanyu Yan, Yan Wang, Hanzi Huawei Technologies Hangzhou310051 China Ministry of Education of China Xiamen University Fujian Key Laboratory of Sensing and Computing for Smart City School of Informatics Key Laboratory of Multimedia Trusted Perception and Efficient Computing Xiamen361005 China Shanghai Artificial Intelligence Laboratory Shanghai200232 China

Existing state-of-the-art methods for few-shot action recognition (FSAR) achieve promising performance by spatial and temporal modeling. However, most current methods ignore the importance of edge information and motion cues, leading to inferior performance. For the few-shot task, it is important to effectively explore limited data. Additionally, effectively utilizing edge information is beneficial for exploring motion cues, and vice versa. In this paper, we propose a novel edge guided network with motion enhancement (EGME) for FSAR. To the best of our knowledge, this is the first work to utilize the edge information as guidance in the FSAR task. Our EGME contains two crucial components, including an edge information extractor (EIE) and a motion enhancement module (ME). Specifically, EIE is used to obtain edge information on video frames. Afterward, the edge information is used as guidance to fuse with the frame features. In addition, ME can adaptively capture motion-sensitive features of videos. It adopts a self-gating mechanism to highlight motion-sensitive regions in videos from a large temporal receptive field. Based on the above designed components, EGME can capture edge information and motion cues, resulting in superior recognition performance. Experimental results on four challenging benchmarks show that EGME performs favorably against recent advanced methods. © 1991-2012 IEEE.

关键词： Motion capture

来源：评论

学校读者我要写书评

暂无评论

Leveraging Machine Learning for Threat Detection: A Study on Malicious URL Classification

Leveraging Machine Learning for Threat Detection: A Study on...

引用

2024 IEEE International Conference on computing, ICOCO 2024

作者： Sheng, Justin Wong Junn Mathivanan, Norsyela Muhammad Noor Ghani, Nor Azura Md School of Computing and Creative Media University of Wollongong Malaysia Shah Alam Malaysia School of Computing and Creative Media University of Wollongong Malaysia College of Computing Informatics and Mathematics Universiti Teknologi MARA Shah Alam Malaysia

ISBN: (纸本)9798331530303

In today's digital age, the internet serves as a vital resource. However, it poses significant risks from malicious websites that threaten cybersecurity. This study focuses on developing effective threat detection methods to identify harmful URLs using machine learning models, specifically Random Forest and XGBoost, to classify URL types such as benign, phishing, malware, and defacement. A comprehensive feature extraction process trains the classification models for real-time analysis of URL threats. The findings indicate that the Random Forest model outperforms XGBoost, particularly in classifying phishing URLs, while demonstrating robust accuracy despite challenges with overfitting. This research culminates in the creation of an Android-based Malicious URL Detector mobile application that integrates the trained model, aiming to enhance cybersecurity defenses. By leveraging machine learning techniques, this study contributes to more efficient and effective detection of malicious URLs, fostering a safer online environment for users. ©2024 IEEE.

关键词： Phishing

来源：评论

学校读者我要写书评

暂无评论

Pneumonia Detection System Using Convolutional Neural Network 5

Pneumonia Detection System Using Convolutional Neural Networ...

引用

5th International Conference on Artificial Intelligence and Data Sciences, AiDAS 2024

作者： Izzat, Muhammad Fazrul Sa'dan, Siti 'Aisyah Bahrin, Ummu Fatihah Mohd Ishak, Siti Nurul Hatie Yasin, Sharifah Nurulhikmah Syed Ibrahim, Normalina Nor, Mat College of Computing Informatics and Media Universiti Teknologi Mara Terengganu Branch Kuala Terengganu Campus Terengganu Kuala Terengganu21080 Malaysia College of Computing Informatics and Media Universiti Teknologi MARA Pahang Branch Raub Campus Pahang Raub27600 Malaysia

ISBN: (纸本)9798331528553

Pneumonia stands as a critical respiratory infection, exerting a significant impact on a global scale. With its ability to affect millions of individuals across various age groups and regions, the urgency surrounding this medical concern becomes clear. Early and accurate diagnosis emerges as a crucial factor in effectively addressing this health challenge. Timely identification of pneumonia enables medical professionals to promptly initiate appropriate treatments, thereby minimizing potential complications and improving patient outcomes. Given the potential severity of the infection, the importance of swift and precise diagnosis cannot be overstated, underscoring the vital role it plays in combating the spread and impact of this respiratory ailment. This study presents a pneumonia detection system to aid in timely diagnosis based on chest X-ray images using the Convolutional Neural Network (CNN) algorithm. The study demonstrates impressive performance with F1-Score of 95.82% and accuracy of 95.19%, demonstrating high reliability and accuracy. The system prototype integrates the trained model with Flask as the back-end and HTML, CSS, and JavaScript for the front-end, providing users with real-time predictions and confidence scores. By leveraging the capabilities of CNN and a user-friendly interface, this pneumonia detection system enables medical professionals to make informed decisions, potentially leading to better treatment outcomes. It also presents as a valuable tool for healthcare practices. © 2024 IEEE.

关键词： Chest X-ray Images Convolutional Neural Network Image Detection Pneumonia

来源：评论

学校读者我要写书评

暂无评论

Performance Evaluation of the Fourth-Generation Xeon with Different Memory Characteristics

Performance Evaluation of the Fourth-Generation Xeon with Di...

引用

2024 International Conference on High Performance computing in Asia-Pacific Region Workshops, HPC Asia 2024 Workshops

作者： Fukazawa, Keiichiro Takahashi, Riki Academic Center for Computing and Media Studies Kyoto University Japan Graduate School of Informatics Kyoto University Japan

ISBN: (纸本)9798400716522

At the Supercomputer System of Academic Center for computing and Media Studies Kyoto University, the fourth-generation Xeon (code-named Sapphire Rapids) is employed. The system consists of two subsystems—one equipped solely with high-bandwidth memory, HBM2e, and the other with a large DDR5 memory capacity. Using benchmark applications, a performance evaluation of systems with each type of memory was conducted. Additionally, the study employed a real application, the electromagnetic fluid code, to investigate how application performance varies based on differences in memory characteristics. The results confirm the performance improvement due to the high bandwidth of HBM2e. However, it was also observed that the efficiency is lower when using HBM2e, and the effects of cache memory optimization are relatively minimal. © 2024 Copyright held by the owner/author(s).

关键词： Sapphire

来源：评论

学校读者我要写书评

暂无评论

Sustainable energy management in the AI era: a comprehensive analysis of ML and DL approaches

引用

computing 2025年第6期107卷 1-64页

作者： Javed, Haseeb Eid, Fatma El-Sappagh, Shaker Abuhmed, Tamer Department of Computer Science and Engineering College of Computing and Informatics Sungkyunkwan University Suwon Korea Republic of Technology Management Stony Brook University NY11794 United States Applied Artificial Intelligence College of Computing and Informatics Sungkyunkwan University Suwon Korea Republic of Faculty of Computer Science and Engineering Galala University Suez435611 Egypt Information Systems Department Faculty of Computers and Artificial Intelligence Benha University Banha13518 Egypt

This study comprehensively analyzes the application of innovative deep learning (DL) and machine learning (ML) techniques in smart energy management systems (EMSs), with an emphasis on load forecasting, demand response, and the development of smart energy sectors. The application of various ML and DL models were examined in over 200 studies from 2014 to 2024 in an electrical network's EMS to highlight the key benefits and advances made by each technology for the sustainable management systems in energy sector. The findings emphasize DL and ML models’ enhanced precision and predictive capabilities in load forecasting, their efficacy in enabling efficient demand response mechanisms, and their significance in supporting the development of smart energy sectors. Furthermore, recommendations are made based on the survey results to assist in incorporating these techniques into EMS frameworks, such as investment in data infrastructure, model training and validation, and collaboration between researchers, industry experts, and policymakers. The study also discusses the limitations identified in the literature, such as limited real-world implementations, challenges regarding quality and data availability, and the need for enhanced ML and DL model interpretability. Addressing these limitations can assist in increasing the application and efficacy of ML and DL techniques in EMSs, enabling a more efficient and sustainable energy landscape. Finally, this study facilitates researchers' exploration of ML and DL in energy management, highlighting relevant limitations, strengths, and alternative approaches associated with sustainable energy management. It also indicates potential future research directions for further investigation. © The Author(s) 2025.

关键词： Power management (telecommunication)

来源：评论

学校读者我要写书评

暂无评论

Natural Language Processing Revolution From Humble Beginnings to Deep Learning and Beyond 4

Natural Language Processing Revolution From Humble Beginning...

引用

4th IEEE International Conference on ICT in Business Industry and Government, ICTBIG 2024

作者： Smarts, Nyalalani Selvaraj, Rajalakshmi Kuthadi, Venumadhav Botswana International University of Science and Technology School of Pure and Applied Sciences Department of Computing and Informatics Palapye Botswana

ISBN: (纸本)9798331518981

This paper explores the intricate development of Natural Language Processing (NLP), a domain which is centered at the core of computational intelligence and human communication. This research is inspired by the need to understand the foundations of NLP and how it has influenced the development of human-computer interactions, language comprehension systems and information retrieval techniques. It explores the transformational effect of deep learning and attention-based models such as the third-generation Generative Pre-trained Transformer (GPT-3), Text-to-Text Transfer Transformer (T5) and Bidirectional Encoder Representations from Transformers (BERT). The models mentioned above have greatly improved the capability to comprehend and generate human-like language using computers. The originality of the work is also reflected on its comprehensive overview of key NLP concepts, tracing chronological advancements, exploring deep learning innovations, and discussing effective pruning techniques to optimize model performance. By reviewing the history of NLP, this research not merely brings the past into focus but suggests future developments for scholars to investigate and find solutions for current challenges in NLP. Its results not only act as a guide for practitioners today but also suggest new directions for research and development that will help meet the challenges of tomorrow in this dynamic discipline of NLP. © 2024 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

COMBINING HAND POSE ESTIMATION WITH AUDIO FOR NOISE-ROBUST PIANO SCORE FOLLOWING 21

COMBINING HAND POSE ESTIMATION WITH AUDIO FOR NOISE-ROBUST P...

引用

21st Sound and Music computing Conference, SMC 2024

作者： Pet, Kaitlin Maezawa, Akira Luddy School of Informatics Computing and Engineering Indiana University BloomingtonIN United States Yamaha Corpration Hamamatsu Japan

ISBN: (纸本)9789893520758

Tracking the musical score position from a user's piano performance is difficult in noisy environments and other cases when the audio signal is unreliable. We present a multi-modal score following method that seamlessly fuses audio and hand tracking from top-down video of a pianist's hands on the keyboard, using a probabilistic model and a feature-weighing mechanism. Experiments demonstrate our method's robustness under noisy environments and when pieces contain mistakes, compared to a score follower based only on audio input. © 2024. This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 Unported License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original.

关键词： Palmprint recognition

来源：评论

学校读者我要写书评

暂无评论

Extreme Learning Machine Optimization for Employee Satisfaction With Modified Metaheuristic 4

Extreme Learning Machine Optimization for Employee Satisfact...

引用

4th International Conference on Expert Clouds and Applications, ICOECA 2024

作者： Abdulla, Tasnim Jovanovic, Luka Radomirovic, Branislav Njegus, Angelina Zivkovic, Miodrag Bacanin, Nebojsa Singidunum University Faculty of Informatics and Computing Belgrade Serbia Singidunum University Faculty of Technical Sciences Belgrade Serbia

ISBN: (纸本)9798350385793

In the contemporary business landscape, the success of a company is intricately linked to the engagement and satisfaction of its workforce. This study analyzes the signifi-cance of developing a contented and engaged employee base, emphasizing the direct impact of workplace satisfaction on productivity, turnover rates, and overall organizational dynamics. Organizational culture emerges as a pivotal factor influencing the recruitment, retention, and satisfaction of talented employees. To address the complexities of identifying and mitigating employee dissatisfaction, this research work proposes a comprehensive solution harnessing the capabilities of cloud-based technologies, specifically text mining, Natural Language Processing (NLP), and modified metaheuristic techniques. The study explores the application of an extreme learning machine as a classifier for assessing employee satisfaction within a cloud computing framework. Acknowledging the critical role of hyperparameter selection in model performance, metaheuristic optimizers and cloud platforms implementation are employed to enhance accu-racy and effectiveness. Furthermore, a novel modification to a metaheuristic algorithm for satisfying the unique requirements of this research is introduced. This research study demonstrates the efficiency of the optimized models, achieving an accuracy rate surpassing 84%. By integrating cloud computing technologies into the proposed framework, organizations gain a powerful and scalable tool for proactively identifying and addressing employee dissatisfaction, ultimately contributing to the improved employee well-being and organizational success in the cloud era. © 2024 IEEE.

关键词： Job satisfaction

来源：评论

学校读者我要写书评

暂无评论

GAProtoNet: A Multi-head Graph Attention-based Prototypical Network for Interpretable Text Classification 31

GAProtoNet: A Multi-head Graph Attention-based Prototypical ...

引用

31st International Conference on Computational Linguistics, COLING 2025

作者： Wen, Ximing Tan, Wenjuan Weber, Rosina O. College of Computing and Informatics Drexel University Philadelphia United States Department of Computer Science Tsinghua University Beijing China

ISBN: (纸本)9798891761964

Pretrained transformer-based Language Models (LMs) are well-known for their ability to achieve significant improvement on text classification tasks with their powerful word embeddings, but their black-box nature, which leads to a lack of interpretability, has been a major concern. In this work, we introduce GAProtoNet, a novel white-box Multi-head Graph Attention-based Prototypical Network designed to explain the decisions of text classification models built with LM encoders. In our approach, the input vector and prototypes are regarded as nodes within a graph, and we utilize multi-head graph attention to selectively construct edges between the input node and prototype nodes to learn an interpretable prototypical representation. During inference, the model makes decisions based on a linear combination of activated prototypes weighted by the attention score assigned for each prototype, allowing its choices to be transparently explained by the attention weights and the prototypes. Experiments on multiple public datasets show our approach achieves superior results without sacrificing the accuracy of the original black-box LMs. We also compare with four alternative prototypical network variations and our approach achieves the best accuracy and F1 among all. Our case study and visualization of prototype clusters also demonstrate the efficiency in explaining the decisions of black-box models built with LMs. Our codes are available at https://***/ximingwen/GAProtoNet. © 2025 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Exploring Lightweight YOLOv8 Architectures for Circuit Board Defect Detection

Exploring Lightweight YOLOv8 Architectures for Circuit Board...

引用

2024 IEEE International Conference on Intelligent computing and Emerging Communication Technologies, ICEC 2024

作者： Jovanovic, Luka Jokic, Aleksandar Milovanovic, Marina Antonijevic, Milos Zivkovic, Miodrag Bacanin, Nebojsa Singidunum University Faculty of Technical Sciences Belgrade Serbia Singidunum University Faculty of Informatics and Computing Belgrade Serbia

ISBN: (纸本)9798331508432

Modern electronics relies heavily on circuit boards, which must be designed and produced using a variety of procedures in order to be filled with electrical components. This procedure includes drilling and plating holes, applying masks, painting and etching fiberglass-reinforced epoxy resin boards, and adding components to the boards. Because of variations in the materials and processes, mistakes might occur at any stage. Traditionally, conductivity, impedance, and visual testing are used in a multi-stage approach to identify and resolve these problems. With modest hardware requirements, computer vision, especially using convolutional networks, provides considerable promise to improve defect detection. Once implemented, these models may significantly increase error detection and lower board failure rates, despite the significant hardware requirements for training. Using a publicly accessible dataset, this study investigates the use of the YOLO model for fault identification in circuit board fabrication. The YOLOv8 model's lightweight design is examined;the medium-sized model achieves a mAP@50 score of 0.990, indicating that it considerably improves fault detection. © 2024 IEEE.

关键词： Digital storage

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：