检索结果-内蒙古大学图书馆

Algorithm for Mining Maximal Balanced Bicliques Using Formal Concept Analysis

IEEE ACCESS 2025年 13卷 35113-35123页

作者： Sadriddinov, Ilkhomjon Peng, Sony Siet, Sophort Kim, Dae-Young Park, Doo-Soon Yi, Gangman Soonchunhyang Univ Dept Software Convergence Asan 31538 South Korea Soonchunhyang Univ Dept Comp Sci Engn Asan 31538 South Korea Dongguk Univ Dept AI Software Convergence Seoul 04260 South Korea

One of the most fundamental models for cohesive subgraph mining in network analysis is that which involves the use of cliques. In bipartite graph analysis, the detection of maximal balanced bicliques (MBB) is an important problem with numerous applications, including VLSI design, protein interactions, and social networks. However, MBB detection is difficult, complex, and time-consuming. In the current paper, to address these disadvantages, we propose a new algorithm for detecting MBB using formal concept analysis (FCA) on bipartite graphs. We applied an algorithm to compute formal concepts from the formal context, which is an alternative way of representing a bipartite graph. We proved that the MBB problem is equivalent to the semi-equiconcept enumeration problem in the formal context. Therefore, the semi-equiconcept mining algorithm was applied to the MBB enumeration problem. However, since the existing FCA algorithm cannot be directly applied to mine all MBBs, the FCA algorithm was modified in its entirety. Thorough asymptotic analysis was performed on the proposed algorithm. Experiments were also conducted on various real-world bipartite graphs to which the proposed algorithm was applied, and our results were found to be significantly better than those obtained by the preexisting algorithm.

关键词： Bipartite graph Formal concept analysis Lattices Web pages Time complexity Social networking (online) detection algorithms Data mining formal concept analysis maximal balanced biclique

来源：评论

学校读者我要写书评

暂无评论

Fixed-Time Bipartite Containment Control for Heterogeneous Multiagent Systems Under DoS Attacks: An Event-Triggered Mechanism

引用

IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS 2025年第4期55卷 2782-2794页

作者： Wang, Li Yan, Huaicheng Hu, Xiao Li, Zhichen Wang, Meng East China Univ Sci & Technol Key Lab Smart Mfg Energy Chem Proc Minist Educ Shanghai 200237 Peoples R China East China Univ Sci & Technol Key Lab Smart Mfg Energy Chem Proc Minist Educ Shanghai 200237 Peoples R China Weifang Univ Coll Machinery & Automat Weifang 261061 Peoples R China Xian Univ Technol Sch Elect Engn Xian 710048 Peoples R China

This article investigates secure bipartite containment control with the event-triggered mechanism (ETM) for nonlinear heterogeneous multiagent systems (MASs) via the feedback control in fixed time. A novel attacks detection algorithm and an updated label strategy are designed to judge the occurrence of Denial of Service (DoS) attacks or not. It can search for new multiple directed spanning trees for MASs with multiple leaders to decrease the negative influence when attacks happen. In this case, the dynamic ETM is adopted by adjusting the triggered threshold online to save resource consumption. The state-feedback control and output-feedback control are selectively employed according to the situation where the state value of the follower is available or not. The settling time is further relaxed by fixed-time stability theory and can be preset in advance. By theoretical discussion, conditions of achieving bipartite containment control are derived by Lyapunov functions. Finally, the effectiveness of the established control method is verified by simulations.

关键词： Event detection detection algorithms Switches Multi-agent systems Topology Control systems Vehicle dynamics Stability criteria Asymptotic stability Smart manufacturing Denial of Service (DoS) attacks event-triggered control feedback control fixed-time stability nonlinear heterogeneous multiagent systems (MASs)

来源：评论

学校读者我要写书评

暂无评论

Classification of Oral Cancer Into Pre-Cancerous Stages From White Light Images Using LightGBM Algorithm

引用

IEEE ACCESS 2024年 12卷 31626-31639页

作者： Goswami, Bibek Bhuyan, M. K. Alfarhood, Sultan Safran, Mejdl Indian Inst Technol Guwahati Dept Elect & Elect Engn Gauhati 781039 India King Saud Univ Coll Comp & Informat Sci Dept Comp Sci Riyadh 11543 Saudi Arabia

Cancer is one of the foremost reasons for death worldwide, with nearly 10 million deaths noted in 2020. Globally, oral cancer ranks sixth when compared to other cancers. It is lethal because most cases are noticed at advanced stages, which can be prevented if screened for or treated early in the pre-cancerous stages, successively leading to a significant decrease in the mortality rate. In this work, a method is proposed that can effectively differentiate between benign and malignant oral cavity lesions and also classify their pre-cancerous stages. The method involves exploring five distinct color spaces and extracting color and texture features, which are then classified using a machine learning technique called Light Gradient Boosting Machine (LightGBM). The overall performance is promising, outperforming the state-of-art methods for the task of oral cancer classification, with a testing accuracy of 99.25%, precision of 99.18%, recall of 99.31%, f1-score of 99.24% and specificity of 99.31% for the binary classification, and testing accuracy of 98.88%, precision of 98.86%, recall of 97.92%, f1-score of 98.38% and specificity of 99.03% for multi-class classification. The proposed method used hand-crafted features and a machine-learning classifier, which uses limited resources and is less time-consuming.

关键词： Image color analysis Cancer Lesions Feature extraction Tongue Classification algorithms Image segmentation detection algorithms Biomedical imaging Binary and multi-class classification color spaces early detection feature importance oral cancer white light images

来源：评论

学校读者我要写书评

暂无评论

Low-Power Real-Time Seizure Monitoring Using AI-Assisted Sonification of Neonatal EEG

引用

IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTING 2025年第1期13卷 80-89页

作者： Nguyen, Tien Daly, Aengus Gomez-Quintana, Sergi O'Sullivan, Feargal Temko, Andriy Popovici, Emanuel Univ Coll Cork Dept Elect & Elect Engn Coll Rd Cork T12 K8AF Ireland

Detecting seizures in neonates requires continuous electroencephalography (EEG) monitoring, a costly process that demands trained experts. Although recent advancements in machine learning offer promising solutions for automated seizure detection, the opaque nature of these algorithms poses significant challenges to their adoption in healthcare settings. A prior study demonstrated that integrating machine learning with sonification-an interpretation method that converts bio-signals into sound-can mitigate the black-box problem while enhancing seizure detection performance. This AI-assisted sonification algorithm can provide a valuable complementary tool in seizure monitoring besides the traditional visualization method. A low-power and affordable implementation of the algorithm is presented in this study using a microcontroller. To improve its practicality, we also introduce a real-time design that allows the sonification algorithm to function in parallel with data acquisition. The system consumes 12 mW in average, making it suitable for a battery-powered device.

关键词： Electroencephalography Sonification Pediatrics Real-time systems Instruction sets Signal processing algorithms Vocoders Monitoring Interpolation detection algorithms Artificial intelligence biomedical monitoring biomedical signal processing electroencephalography embedded software Edge AI internet of medical things machine learning microprocessors real-time systems sonification signal processing algorithms

来源：评论

学校读者我要写书评

暂无评论

A Privacy-Preserving Collaborative Federated Learning Framework for Detecting Retinal Diseases

引用

IEEE ACCESS 2024年 12卷 170176-170203页

作者： Gulati, Seema Guleria, Kalpna Goyal, Nitin Alzubi, Ahmad Ali Castilla, Angel Kuc Chitkara Univ Inst Engn & Technol Chandigarh 140401 Punjab India Cent Univ Haryana Sch Engn & Technol Dept Comp Sci & Engn Mahendergarh 123031 Haryana India King Saud Univ Community Coll Comp Sci Dept Riyadh Saudi Arabia Univ Europea Atlantico Engn Res & Innovat Grp Santander 39011 Spain Univ Int Iberoamer Dept Project Management Campeche 24560 Mexico Univ Int Iberoamer Dept Project Management Arecibo PR 00613 USA

The rapid advancement in technology has simplified human life and provides convenience. However, this convenience has led to many lifestyle diseases like diabetes and obesity. The 2021 reports of the International Diabetes Federation (IDF) show 537 million diabetics, three-fourths of whom are from developing countries. About 28.5% of these diabetics over 40 suffer from Diabetic Retinopathy (DR), 4.4% face vision-threatening DR, and 3.8% suffer from Diabetic Macular Edema (DME). These conditions can lead to complete vision loss affecting health and quality of life. Early detection of DR and DME is crucial to prevent harmful effects. The proposed work employs a collaborative, privacy-preserving Federated Deep Learning (FDL) framework with lightweight MobileNetV2 architecture for early detection of DR and DME. The proposed FDL framework uses both Independent and Identically Distributed (IID) and non-IID data for 2 and 3-client architectures. In a 2-client scenario, the FDL implementation with FedAvg aggregation achieved 98.69% accuracy on IID data and 87.09% on non-IID data, while FedProx aggregation scored 98.03% on IID data and 98.25% on non-IID data. In a 3-client scenario, FedAvg aggregation achieved 97.62% accuracy on IID data and 96.28% on non-IID data, whereas FedProx aggregation achieved 98.69% on IID data and 97.77% on non-IID data. The results demonstrate that the FedProx aggregation is more stable and converges earlier than FedAvg aggregation in the non-IID settings of an FDL framework. The proposed FDL framework with its collaborative training feature, preserves privacy and maintains high prediction accuracy.

关键词： Training Data models Accuracy Data privacy Collaboration Servers Medical services Deep learning Computational modeling Eye diseases detection algorithms Federated learning Machine learning Automatic disease detection early convergence federated deep learning framework healthcare machine learning MobileNetV2 model stability retinal diseases

来源：评论

学校读者我要写书评

暂无评论

CHEAT: A Large-Scale Dataset for Detecting CHatGPT-writtEn AbsTracts

引用

IEEE TRANSACTIONS ON BIG DATA 2025年第3期11卷 898-906页

作者： Yu, Peipeng Chen, Jiahan Feng, Xuan Xia, Zhihua Jinan Univ Coll Cyber Secur Engn Res Ctr Trustworthy AI Minist Educ Guangzhou 510632 Peoples R China

The powerful ability of ChatGPT has caused widespread concern in the academic community. Malicious users could synthesize dummy academic content through ChatGPT, which is extremely harmful to academic rigor and originality. The need to develop ChatGPT-written content detection algorithms calls for large-scale datasets. In this paper, we initially investigate the possible negative impact of ChatGPT on academia, and present a large-scale CHatGPT-writtEn AbsTract dataset (CHEAT) to support the development of detection algorithms. In particular, the ChatGPT-written abstract dataset contains 35,304 synthetic abstracts, with $Generation$Generation, $Polish$Polish, and $Fusion$Fusion as prominent representatives. Based on these data, we perform a thorough analysis of the existing text synthesis detection algorithms. We show that ChatGPT-written abstracts are detectable with well-trained detectors, while the detection difficulty increases with more human guidance involved.

关键词： Chatbots Feature extraction detection algorithms Big Data Social networking (online) Detectors Reinforcement learning Oral communication Deep learning Accuracy ChatGPT synthetic abstracts text synthesis detection

来源：评论

学校读者我要写书评

暂无评论

Synchrony Vision: Capturing Body Motion Synchrony Through Phase Difference Using the Kinect

引用

IEEE ACCESS 2025年 13卷 41658-41669页

作者： Kwon, Jinhwan Kyoto Univ Educ Dept Educ Kyoto 6128522 Japan

This study introduces a Kinect-based system for real-time detection of body motion synchrony, addressing limitations of traditional methods such as high costs, intrusiveness, subjectivity, and lack of real-time analysis. Using a phase difference detection algorithm, the system analyzes synchrony offering a comprehensive framework for quantifying interpersonal coordination. The system captures acceleration data from up to six individuals, calculates phase differences, and provides immediate feedback. The original algorithm was modified to address challenges such as gravitational acceleration, inverted axes, and peak detection thresholds, improving the accuracy of Kinect data analysis. Experimental results demonstrate the system's overall detection accuracy of 89.2% under controlled conditions. Additionally, a comparison with an accelerometer-based method revealed a strong correlation (r = 0.73, p = 0.002), indicating alignment in detecting synchrony. While the Kinect-based system offers advantages in scalability and usability, it exhibits limitations in detecting high-speed movements. To enhance its accuracy, potential improvements include algorithmic refinements, hardware upgrades, and AI-driven models to adaptively refine motion detection.

关键词： Sensors Real-time systems Synchronization detection algorithms Tracking Costs Accuracy Sensor systems Optical sensors Data visualization Body motion synchrony Kinect-based system phase difference detection real-time analysis non-invasive systems

来源：评论

学校读者我要写书评

暂无评论

Real-Time Monitoring of Personal Protective Equipment Compliance in Surveillance Cameras

引用

IEEE ACCESS 2024年 12卷 121882-121895页

作者： Al-Azani, Sadam Luqman, Hamzah Alfarraj, Motaz Sidig, Ala Addin I. Khan, Ayaz H. Al-Hamed, Dina King Fahd Univ Petr & Minerals SDAIA KFUPM Joint Res Ctr Artificial Intelligence Dhahran 31261 Saudi Arabia King Fahd Univ Petr & Minerals Informat & Comp Sci Dept Dhahran 31261 Saudi Arabia King Fahd Univ Petr & Minerals Elect Engn Dept Dhahran 31261 Saudi Arabia King Fahd Univ Petr & Minerals Comp Engn Dept Dhahran 31261 Saudi Arabia

Worker safety in the industrial, construction, and energy sectors is the primary concern of governments, organizations, and project managers. Traditional monitoring systems are tedious, laborious, subjective, expensive, and inaccurate. Therefore, the development of automated and accurate systems for real-time monitoring and detecting personal protective equipment (PPE) compliance is urgently needed. This study provides valuable insights and contributions to the field of workers' safety in the industrial, construction, and energy sectors. It aims to evaluate the potential contribution of providing a robust and accurate real-time PPE compliance monitoring system using state-of-the-art deep learning techniques with real-world videos collected from surveillance cameras. Furthermore, it presents a systematic framework for developing, evaluating, and selecting the most suitable model for real-world applications, starting from data collection and ending with system deployment and maintenance. To the best of our knowledge, the collected dataset in this work is the largest PPE's dataset, with approximately 7.7 million labels representing different PPEs in 386,011 video frames collected from several surveillance cameras. The techniques adopted to generate different models and evaluated for PPE are CenterNet, Vision Transformer, and YOLOv7. Moreover, an optimized real-time inference engine is developed leveraging the high-computing power of GPUs in the system. The experimental results demonstrate that the models developed using YOLO are the most accurate and can be better generalized to detect PPE compliance in new working environments, achieving mean average precisions of 92.36% and 66.70% on Test Set I and Test Set II, respectively. In addition, they achieve real-time detection of PPE compliance.

关键词： Personal protective equipment Safety Head Cameras Real-time systems YOLO Videos detection algorithms Program processors Multithreading PPE monitoring PPE detection multithreading vision transformers objects as points

来源：评论

学校读者我要写书评

暂无评论

Detecting and Analyzing Motifs in Large-Scale Online Transaction Networks

引用

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2025年第2期37卷 584-596页

作者： Jiang, Jiawei Huang, Hao Zheng, Zhigao Wei, Yi Fu, Fangcheng Li, Xiaosen Cui, Bin Wuhan Univ Sch Comp Sci Wuhan 430072 Peoples R China Peking Univ Sch Comp Sci Beijing 100871 Peoples R China Tencen Inc Shenzhen 518054 Peoples R China

Motif detection is a graph algorithm that detects certain local structures in a graph. Although network motif has been studied in graph analytics, e.g., social network and biological network, it is yet unclear whether network motif is useful for analyzing online transaction network that is generated in applications such as instant messaging and e-commerce. In an online transaction network, each vertex represents a user's account and each edge represents a money transaction between two users. In this work, we try to analyze online transaction networks with network motifs. We design motif-based vertex embedding that integrates motif counts and centrality measurements. Furthermore, we design a distributed framework to detect motifs in large-scale online transaction networks. Our framework obtains the edge directions using a bi-directional tagging method and avoids redundant detection with a reduced view of neighboring vertices. We implement the proposed framework under the parameter server architecture. In the evaluation, we analyze different kinds of online transaction networks w.r.t the distribution of motifs and evaluate the effectiveness of motif-based embedding in downstream graph analytical tasks. The experimental results also show that our proposed motif detection framework can efficiently handle large-scale graphs.

关键词： Image edge detection Network motifs Social networking (online) Biology Servers Tagging Proteins Indexes Feature extraction detection algorithms Online transaction network network motif parameter server

来源：评论

学校读者我要写书评

暂无评论

Decoding Student Emotions: An Advanced CNN Approach for Behavior Analysis Application Using Uniform Local Binary Pattern

引用

IEEE ACCESS 2024年 12卷 106273-106284页

作者： Sowjanya, Uppanapalli Lakshmi Krithiga, R. Vellore Inst Technol Sch Comp Sci & Engn Chennai 600127 India

In this paper, Facial Emotion detection model, built using Uniform Local Binary Pattern, is explained using an application i.e., Student behavior detection. Student engagement is a critical factor in effective teaching and learning. Conventional approaches to evaluating student participation frequently depend on assertions made by students and observations made by teachers, which can be laborious and unreliable. The use of computer vision techniques to automatically identify the emotions and behaviors of students in real time has gained popularity in recent years. This research provides a novel computer vision-based method for detecting student behavior, emphasizing emotion recognition. We propose a system that utilizes Convolutional Neural Networks (CNNs) for emotion classification, data augmentation to enhance dataset, and Uniform Local Binary pattern (uLBP) feature extraction for improved texture analysis. The system aims to provide educators with real-time insights into student emotions, enabling them to adapt their teaching strategies accordingly. In order to expand the variety and breadth of the training dataset, we enhanced the facial expression data that we extracted from a dataset that's well suited for this work. We then applied uLBP feature extraction to capture local texture patterns in the facial images, which were then used as input to the CNN for emotion classification. According to our findings, the suggested approach successfully classified student emotions with a high accuracy of 94% and also completed the ablation study, which indicates its usefulness in practical settings.

关键词： Computer vision Deep learning Behavioral sciences Education detection algorithms deep learning emotion detection student behavior data augmentation uniform local binary pattern (uLBP) Education detection algorithms deep learning emotion detection student behavior data augmentation uniform local binary pattern (uLBP)

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：