检索结果-内蒙古大学图书馆

Proceedings of the 2024 International Conference on Image Processing, Multimedia Technology and Maching Learning

作者： Zhiping Zhang Shuzlina Abdul-Rahman Norlina Mohd Sabri School of Computing Sciences College of Computing Informatics and Media Universiti Teknologi MARA Selangor Malaysia College of Computer and Mathematics Xinyu University Xinyu Jiangxi China School of Computing Sciences College of Computing Informatics and Media Universiti Teknologi MARA Selangor Malaysia Universiti Teknologi MARA Cawangan Terengganu Kuala Terengganu Malaysia

ISBN: (纸本)9798400713880

Facial expression recognition is of crucial importance in a variety of applications, including human-computer interaction and health monitoring. However, when there is occlusion, especially when the mouth region is occluded, conventional FER methods often suffer from significant performance degradation. To address this challenge, a novel FER model, Mini-Xception+CBAM, is proposed. This combines a lightweight network architecture with an attention mechanism to enhance performance in occlusion scenarios. The proposed approach utilises the Mini-Xception architecture and integrates a Convolutional Block Attention Module (CBAM) to enable the model to dynamically focus on key facial regions, such as the eyes and eyebrows, which are less likely to be occluded. Extensive experimentation on the FER2013, SMFER2013, and MFER2013 datasets demonstrates that the proposed model outperforms numerous state-of-the-art baseline models, including Mini-Xception, MobileNetV2, ResNet-18, and Vision Transformer, achieving The model achieved 90.4% accuracy on the FER2013, 89.2% on the SMFER2013, and 91.5% on the MFER2013, along with higher precision, recall, and F1 score. Furthermore, the model has been designed to be efficient, with a parameter count of 0.68M and a floating point operations count of 30.1M, making it ideal for deployment on devices with limited resources.

关键词： Expression recognition

来源：评论

学校读者我要写书评

暂无评论

APTO: Accelerating Serialization-Based Point Cloud Transformers with Position-Aware Pruning 25

APTO: Accelerating Serialization-Based Point Cloud Transform...

引用

30th Asia and South Pacific Design Automation Conference, ASP-DAC 2025

作者： Sun, Qichu Meng, Rui Fan, Haishuang Ding, Fangqiang Lu, Linxi Wu, Jingya Li, Xiaowei Yan, Guihai State Key Laboratory of Processors Institute of Computing Technology Chinese Academy of Sciences University of Chinese Academy of Sciences Beijing China School of Informatics University of Edinburgh Edinburgh United Kingdom State Key Laboratory of Processors Institute of Computing Technology Chinese Academy of Sciences Beijing China State Key Laboratory of Processors Institute of Computing Technology Chinese Academy of Sciences Yusur Technology Co. Ltd. Beijing China

ISBN: (纸本)9798400706356

Point cloud processing has broad applications in autonomous driving and robotics. Serialization-based point cloud transformers map unordered point clouds onto directed curves, use sparse convolution for down-sampling and apply attention in local windows to capture spatial relationships. Despite achieving great accuracy, these models face inference latency challenges: neighbor search in sparse convolution exhibits low parallelism;attention computation remains complex, especially with larger window sizes;softmax introduces data dependencies. This paper proposes APTO, an accelerator for serialization-based models. It uses voxels' z-curve indices to perform neighbor searches in parallel, employs a position-aware pruning strategy using neighboring voxel counts to eliminate useless attention computations, and adopts a fine-grained attention dataflow for parallel processes with minimal data dependencies. Besides, its hardware has dedicated computation cores for efficient processing. Evaluations show that APTO achieves average 10.22×, 3.53× and 2.70× speedups over RTX 4090 GPU, PointAcc, and SpOctA, with 153.59×, 8.57× and 7.25× energy savings. © 2025 Institute of Electrical and Electronics Engineers Inc.. All rights reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Leveraging Game Theory and XAI for Data Quality-Driven Sample and Client Selection in Trustworthy Split Federated Learning

引用

IEEE Transactions on Consumer Electronics 2025年

作者： Tariq, Asadullah Sallabi, Farag Serhani, Mohamed Adel Qayyum, Tariq Barka, Ezedin S. United Arab Emirates University College of IT United Arab Emirates College of Computing and Informatics UoS United Arab Emirates

In Federated Learning (FL) systems, clients share updates derived from their local data with the server while maintaining privacy. The server aggregates these updates to refine the global model. However, not all client data may be relevant to the learning objective, and incorporating updates from irrelevant data can harm the model's performance. The selection of training samples significantly impacts model performance, as datasets with errors, skewed distributions, or low diversity can lead to inaccurate and unstable models. To address these issues, in this paper, we proposed a data quality evaluation model to assess the quality of datasets in FL systems. This model aims to dynamically select high-quality data samples for FL training by utilizing intrinsic and contextual data quality dimensions. Additionally, we designed an importance-based interpretable feature selection model and a data quality-based dynamic client selection model that employs Nash equilibrium and joint differential privacy (DP). This approach encourages clients with high-quality data to participate in FL training, thereby improving the overall quality of the training process. We leverage the concept of Split FL to efficiently distribute the model training between the client and server. Our comprehensive approach ensures the selection of high-quality data and features while motivating the participation of reliable clients, ultimately leading to improved performance and stability of FL models. Extensive experimental results show that our proposed model significantly outperforms baseline and comparative schemes. © 1975-2011 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Simple One-step Multi-view Clustering with Fast Similarity and Cluster Structure Learning

引用

IEEE Signal Processing Letters 2025年 32卷 1850-1854页

作者： Xia, Xianxian Huang, Dong Yang, Chen-Min He, Chaobo Wang, Chang-Dong South China Agricultural University College of Mathematics and Informatics Guangzhou China South China Normal University School of Computer Science Guangzhou China Sun Yat-sen University School of Computer Science and Engineering Guangzhou China Ministry of Education Key Laboratory of Machine Intelligence and Advanced Computing China

Multi-view clustering (MVC) is essential for integrating heterogeneous data from multiple sources. However, many existing approaches are hindered by high computational complexity and the separate optimization of similarity and cluster structures. In light of these challenges, this paper presents a novel anchor-based MVC method termed simple one-step multi-view clustering with fast similarity and cluster structure learning (SONIC), which models adaptive anchor learning, multi-view similarity structure learning, and discrete cluster structure learning in a joint framework. In particular, we employ the anchor-based multi-view similarity learning to capture the consensus manifold structure latent in multiple views, thereby constructing a unified bipartite graph with adaptive anchor learning and view weighting. Then we impose a low-rank constraint on the bipartite graph structure to directly reveal the desired number of clusters without additional post-processing. An efficient alternating minimization algorithm is developed to optimize the model, resulting in a computational complexity that scales linearly with the number of samples. Extensive experiments on eight benchmark datasets demonstrate the superior performance of SONIC in both clustering quality and computational efficiency. © 1994-2012 IEEE.

关键词： Data assimilation

来源：评论

学校读者我要写书评

暂无评论

MatrikalinDiabetes: User-Centered Design of a mHealth App for Gestational Diabetes Mellitus Management and Education Among Bangladeshi Women 26th

MatrikalinDiabetes: User-Centered Design of a mHealth App ...

引用

26th International Conference on Human-Computer Interaction, HCII 2024

作者： Ratul, Mohammad Arshad Hossain Yanoor Bristy, Tunisha Sayeed, Noorjahan Islam, Ashraful Chaudhry, Beenish Moalla Center for Computational and Data Sciences Independent University Bangladesh Dhaka Bangladesh Department of Computer Science and Engineering Independent University Bangladesh Dhaka Bangladesh School of Computing and Informatics University of Louisiana at Lafayette Louisiana United States

ISBN: (纸本)9783031768088

This study employs the User-Centered Design (UCD) methodology to develop a mobile health (mHealth) application (app) specifically tailored for Bangladeshi women with Gestational Diabetes Mellitus (GDM). GDM affects approximately 10% of pregnant women globally and around 35% within Bangladesh, as reported by the World Health Organization (WHO). Despite its high prevalence, there is a significant lack of awareness and education about GDM in Bangladesh, compounded by language barriers that make existing digital solutions less accessible. The MatrikalinDiabetes mHealth app aims to overcome these barriers by providing comprehensive management and educational resources for GDM in the Bangla language, addressing both pre-pregnancy and post-pregnancy needs. To ensure the app meets the actual needs of potential users, personas were created based on a literature review, and a survey was conducted. This process informed the development of both a low-fidelity, paper-based prototype and a high-fidelity digital prototype. Key features of the MatrikalinDiabetes app include tracking of food and water intake, physical activity monitoring, reminders, GDM education, chat forums for community support, blood glucose level (BGL) monitoring, an SOS button for emergency contacts, and quizzes for user engagement. Preliminary feedback indicated that 53.8% of participants favored the integrated features and expressed willingness to use and recommend the app. The survey of the high-fidelity prototype yielded positive responses, with 57.9% of participants strongly agreeing and 42.1% agreeing that the features were beneficial. The user-friendly design of the high-fidelity prototype ensures intuitive interaction, aiding in health management, enhancing GDM understanding, and promoting a healthier lifestyle among pregnant women in Bangladesh. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Obstetrics

来源：评论

学校读者我要写书评

暂无评论

Development and Implementation of a Machine Learning-Based Flood Forecasting System in Kasese District, Uganda

引用

Journal of Flood Risk Management 2025年第2期18卷

作者： Miiro, Edward Kato, Ismael Nantege, Zuhra Ssendi, Samuel Bassajjalaba, Khasim Faculty of Computing and Informatics Makerere University Business School Kampala Uganda

This study aimed to develop a proof-of-concept prototype of a machine learning system to forecast and mitigate the effect of floods in Kasese District. The researchers used a participatory design science approach. The researchers conducted document reviews and brainstorming to obtain past climate data from the representatives of affected communities, the Makerere University Department of Meteorology, and the Uganda National Meteorological Authority. Qualitative data were transcribed from recordings of the brainstorming sessions and notes from literature. The data were then summarized in tables and analyzed using Visual Network Analysis (VNA) with Word Clouds and Gephi Open Source Software. We employed a combination of C++ programming, sensors wired to Arduino 2 and 3 Integrated Development Environment System to build the prototype. Two machine learning algorithms, including linear regression and K-nearest neighbours (KNN) were used to learn from collected hydrological data and make necessary predictions. Using sensors, we were able to read water levels, temperature, and humidity. The prototype successfully demonstrated the ability to send early-warning alerts to users, contributing to both theoretical advancements in disaster risk reduction and practical tools for mitigating flood-related losses in Uganda. The researchers recommend further study to validate the use of this system and evaluate its efficacy and predictive accuracy in averting floods in affected areas. © 2025 The Author(s). Journal of Flood Risk Management published by Chartered Institution of Water and Environmental Management and John Wiley & Sons Ltd.

关键词： early warning systems forecasting and flood mitigation Kasese District machine learning

来源：评论

学校读者我要写书评

暂无评论

Toward a Cognitive Data Model: Exploring a Mind-Inspired Approach to Database Design

arXiv

引用

arXiv 2025年

作者： Pieris, Dhammika Department of Information Systems Engineering and Informatics Faculty of Computing University of Sri Jayawardenepura Sri Lanka

The Cognitive Data Model (CDM) is proposed. A novel approach to database design, inspired by the belief that the human brain operates with a logical data model independent of its anatomical structure. The study aims to identify and replicate this CDM to enhance database design, resolving limitations of the existing models in handling complex relationships, scalability, and adaptability. The methodology involves empirical observation, cognitive experiments, iterative modelling and critical thinking. Findings suggest that the information processing in the brain occurs sequentially with forming associations on atomic static data in a pairwise, time-dependent manner. This insight led to the development of a meta-framework for the brain’s information processing to design CDM based on it. The CDM offers improved data modelling and database design, efficient querying, and potential applications in AI, machine learning, and big data analytics. Future research will focus on formalizing the model, implementing it in 3D environments, and developing a query language. Copyright © 2025, The Authors. All rights reserved.

关键词： Structured Query Language

来源：评论

学校读者我要写书评

暂无评论

"Socially Assistive Robot Privacy Model": A Multi-model Approach to Evaluating Socially Assistive Robot Privacy Concerns 16th

"Socially Assistive Robot Privacy Model": A Multi-model Appr...

引用

16th International Conference on Social Robotics, ICSR + AI 2024

作者： Collins, Sawyer Stanojević, Čedomir Bennett, Casey Henkel, Zachary Henkel, Kenna Baugus Abbott, Nikki M. Bethel, Cindy L. Śabanović, Selma Luddy School of Informatics Computing and Engineering Indiana University Bloomington BloomingtonIN United States College of Behavioral Social and Health Sciences Clemson University ClemsonSC United States College of Computing and Digital Media DePaul University ChicagoIL United States Department of Computer Science and Engineering Mississippi State University StarkvilleMS United States

ISBN: (纸本)9789819635214

As socially assistive robots (SARs) enter more diverse care settings, including users’ homes, it is critical to identify the shifting privacy risks and dimensions of privacy this technology and its data collection capabilities may affect. We propose a new model of privacy to address the complex nature of SARs as a multimodal technology within the healthcare space. To construct this new model, we combine a previous three-dimensional model from the healthcare literature and synthesize it with a seven-dimension technology-related model. We then use this new combined model to analyze self-report data from several workshops with prospective users of the dog-like robot Therabot to map out the dimensions of privacy identified as future concerns for clinicians and those living with depression. Finally, we suggest this model can be used in future studies to support the in-depth exploration of privacy implications of SARs within healthcare through discussions about privacy among clinicians, those receiving care, and robot designers. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

The Transparency Challenge in Blockchain-Enabled Sustainable Development Goals Applications: Exploring Privacy-Preserving Techniques and Emerging Platforms

引用

IEEE Access 2025年 13卷 81769-81793页

作者： Ahcène Bounceur Ahmed-Sami Berkani Hamouma Moumen Saber Benharzallah College of Computing and Informatics University of Sharjah Sharjah United Arab Emirates Computer Science Department University of Batna 2 Batna Algeria

Blockchain technology has emerged as a key enabler for achieving the United Nations’ Sustainable Development Goals (SDGs) by offering transparent, decentralized solutions across multiple sectors. However, its inherent transparency introduces significant privacy challenges, particularly in sensitive areas such as healthcare, energy, finance, and supply chains, where confidentiality is critical. To address this tension, this paper explores a range of emerging privacy-preserving techniques, including Ring Signatures, Zero-Knowledge Proofs (ZKPs), Secure Multi-Party Computation (SMPC), Trusted Execution Environments (TEEs), and mixers, analyzing their implementation, effectiveness, and limitations. Furthermore, we conduct a detailed technical study of next-generation blockchain platforms that integrate these techniques to balance privacy, scalability, and interoperability. Our contributions include a deep dive into privacy challenges in blockchain applications for SDGs, a comparative analysis of privacy-preserving solutions, a comprehensive evaluation of emerging platforms, and a review of existing research that leverages these privacy techniques in SDG-related domains. This work aims to provide critical insights into how blockchain technology can be effectively adapted to support sustainable development while safeguarding sensitive data.

关键词： Blockchains Privacy Sustainable development Supply chains Data privacy Medical services Smart contracts Scalability Multi-party computation Technological innovation

来源：评论

学校读者我要写书评

暂无评论

Determining the Covering Radius of All Generalized Zetterberg Codes in Odd Characteristic

引用

IEEE Transactions on Information Theory 2025年第5期71卷 3602-3613页

作者： Shi, Minjia Li, Shitao Helleseth, Tor Ozbudak, Ferruh Anhui University Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education School of Mathematical Sciences Hefei230601 China Anhui University School of Mathematical Sciences Anhui Hefei230601 China University of Bergen Department of Informatics Bergen5007 Norway Sabancı University Faculty of Engineering and Natural Sciences İstanbul34956 Turkey

For an integer s ≥ 1, let Cs(q0) be the generalized Zetterberg code of length q0s + 1 over the finite field Fq0 of odd characteristic. Recently, Shi et al. determined the covering radius of Cs(q0) for (Formula presented) (mod 8), and left the remaining case as an open problem. In this paper, we develop a general technique involving arithmetic of finite fields and algebraic curves over finite fields to determine the covering radius of all generalized Zetterberg codes for q0s ≡ 7 (mod 8), which therefore solves this open problem. We also introduce the concept of twisted half generalized Zetterberg codes of length (Formula presented) , and show the same results hold for them. As a result, we obtain some quasi-perfect codes. © 1963-2012 IEEE.

关键词： Turbo codes

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：