检索结果-内蒙古大学图书馆

Skin cancer detection with MobileNet-based transfer learning and MixNets for enhanced diagnosis

Neural Computing and Applications 2024年第34期36卷 21383-21413页

作者： Zakariah, Mohammed Al-Razgan, Muna Alfakih, Taha Department of Computer Sciences and Engineering College of Applied Science King Saud University Riyadh11543 Saudi Arabia Department of Software Engineering College of Computer and Information Sciences King Saud University Riyadh11345 Saudi Arabia Department of Information Systems College of Computer and Information Sciences King Saud University Riyadh11543 Saudi Arabia

Skin cancer poses a significant health hazard, necessitating the utilization of advanced diagnostic methodologies to facilitate timely detection, owing to its escalating prevalence in recent years. This paper proposes a novel approach to tackle the issue by introducing a method for detecting skin cancer that uses MixNets to enhance diagnosis and leverages mobile network-based transfer learning. Skin cancer has diverse forms, each distinguishable by its structural attributes, morphological characteristics, texture, and coloration. The pressing demand for accurate and efficient diagnostic instruments has spurred the investigation of novel techniques. The present study utilizes the ISIC dataset, comprising a validation set of 660 images and a training set of 2637 images. Moreover, the research employs a combination of MixNets and mobile network-based transfer learning as its chosen approach. Transfer learning is a technique that leverages preexisting models to enhance the diagnostic capabilities of the proposed system. Integrating MobileNet and MixNets allows for utilizing their respective functionalities, resulting in a dual-model methodology that enhances the comprehensiveness of skin cancer diagnosis. The results demonstrate impressive performance metrics, with MobileNet and MixNets models, and the proposed approach achieves an outstanding accuracy rate of 99.58%. The above findings underscore the efficacy of the dual-model method in effectively discerning between benign and malignant skin lesions. Moreover, the present study aims to examine the potential integration of emerging technologies to enhance the accuracy and practicality of diagnostics within real-world healthcare settings. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Diseases

来源：评论

学校读者我要写书评

暂无评论

A Cross Attention Transformer-Mixed Feedback Video Recommendation Algorithm Based on DIEN

引用

computers, Materials & Continua 2025年第1期82卷 977-996页

作者： Jianwei Zhang Zhishang Zhao Zengyu Cai Yuan Feng Liang Zhu Yahui Sun College of Software Engineering Zhengzhou University of Light IndustryZhengzhou450000China Faculty of Information Engineering Xuchang Vocational Technical CollegeXuchang461000China College of Computer Science and Technology Zhengzhou University of Light IndustryZhengzhou450000China College of Electronics and Information Zhengzhou University of Light IndustryZhengzhou450000China

The rapid development of short video platforms poses new challenges for traditional recommendation *** systems typically depend on two types of user behavior feedback to construct user interest profiles:explicit feedback(interactive behavior),which significantly influences users’short-term interests,and implicit feedback(viewing time),which substantially affects their long-term ***,the previous model fails to distinguish between these two feedback methods,leading it to predict only the overall preferences of users based on extensive historical behavior ***,it cannot differentiate between users’long-term and shortterm interests,resulting in low accuracy in describing users’interest states and predicting the evolution of their *** paper introduces a video recommendationmodel calledCAT-MFRec(CrossAttention Transformer-Mixed Feedback Recommendation)designed to differentiate between explicit and implicit user feedback within the DIEN(Deep Interest Evolution Network)*** study emphasizes the separate learning of the two types of behavioral feedback,effectively integrating them through the cross-attention ***,it leverages the long sequence dependence capabilities of Transformer technology to accurately construct user interest profiles and predict the evolution of user *** results indicate that CAT-MF Rec significantly outperforms existing recommendation methods across various performance *** advancement offers new theoretical and practical insights for the development of video recommendations,particularly in addressing complex and dynamic user behavior patterns.

关键词： Video recommendation user interest cross-attention transformer

来源：评论

学校读者我要写书评

暂无评论

A mathematical multi-objective optimization model and metaheuristic algorithm for effective advertising in the social internet of things

引用

Neural Computing and Applications 2025年第6期37卷 4797-4821页

作者： Molaei, Reza Rahsepar Fard, Kheirollah Bouyer, Asgarali Department of Computer Engineering and Information Technology University of Qom Qom Iran Department of Software Engineering Azarbaijan Shahid Madani University Tabriz Iran Department of Software Engineering Faculty of Engineering and Natural Science Istinye University Istanbul Turkey

The integration of social networks with the Internet of Things (IoT) has been explored in recent research, giving rise to the Social Internet of Things (SIoT). One promising application of SIoT is viral marketing, which plays a critical role in modern advertising. However, current solutions proposed for viral marketing are often impractical in real-world scenarios, and there has been limited investigation into viral marketing within the context of SIoT. To address this issue, we present a multi-objective integer linear programming model named ILP-ASoT that facilitates advertising through SIoT. Our approach introduces two realistic parameters: the node sociality rate and the node appropriateness rate as an advertising destination. By balancing the cost of selecting a seed set with the number of users who receive the advertisement, our model offers an appropriate trade-off. Since this is proven as an NP-hard problem, A PSO-based metaheuristic algorithm is proposed to provide near-optimal solutions for advertising in SIoT, called PSO-ASoT. Through simulations, we show that our PSO-ASoT algorithm outperforms state-of-the-art algorithms in several metrics, such as influence spread, advertising cost, and duplicated advertising rate. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Heuristic algorithms

来源：评论

学校读者我要写书评

暂无评论

ENERGY-EFFICIENT TRAFFIC SCHEDULING MECHANISM BASED ON IMPROVED GENETIC ALGORITHM

UPB Scientific Bulletin, Series C: Electrical Engineering an...

引用

UPB Scientific Bulletin, Series C: Electrical engineering and computer Science 2024年第4期86卷 417-436页

作者： Bai, Ningzhuo Li, Honghui Fu, Xueliang College of Computer & Information Engineering Inner Mongolia Agricultural University China

The swift advancements in information technology have ushered in the era of large language models and artificial intelligence. Alongside the rapid expansion of data center scale and business demands, the issue of energy consumption in data center networks has become increasingly prominent. software-Defined Networking (SDN) revolutionizes network architecture by decoupling the data plane from the control plane, enhancing network flexibility and its adoption in data center networks. To tackle data center energy consumption, this paper introduces an energy-saving traffic scheduling algorithm for SDN data center networks, utilizing an enhanced genetic algorithm. Leveraging network monitoring capabilities in SDN, the algorithm enhances traffic scheduling through an advanced genetic algorithm and puts inactive switches into sleep mode to meet the demands of current traffic flow. This approach reduces active network devices, leading to energy savings. Simulation results confirm the method’s feasibility and effectiveness. © 2024, Politechnica University of Bucharest. All rights reserved.

关键词： Data centers

来源：评论

学校读者我要写书评

暂无评论

A study of progressive data flow knowledge tracing based on reconstructed attention mechanism

引用

Neural Computing and Applications 2025年第11期37卷 7675-7689页

作者： Wu, Qianxi Wang, Min Zhou, Guohui Ji, Weidong College of Computer Information Engineering Harbin Normal University Harbin150025 China

Knowledge tracing (KT) is an essential task in intellectual education, which measures learners’ ability to learn new knowledge by collecting historical learning information from learners. With the introduction of Recurrent Neural Networks (RNN) and Transformer into the field of KT, although effective, they focus only on the temporal order in which the learner information is affected. To model accurately, we propose a KT model BPKT (Bayesian-Attention mechanism Progressive data flow for KT) that allows exercise embedding to be layered in different forms and incorporated into the model multiple times. The BPKT model integrates the relationship between exercises covering knowledge points in both the temporal and spatial aspects, and defines a Bayesian-Attention mechanism based on this, with an in-depth analysis of the realistic meaning of the micro-parameters Q, K, and V in the mechanism. Through experiments on four real benchmark datasets, the results show that the BPKT model is helpful for predicting learners’ future responses on large-scale datasets. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Privacy-Preserving Matching Scheme for IoV-Enabled Ride-Sharing Services

引用

IEEE Transactions on Vehicular Technology 2025年第6期74卷 8694-8706页

作者： Zhang, En Huang, Yuchen Qin, Leiyong Henan Normal University College of Computer and Information Engineering Xinxiang453007 China

Ride-sharing services provide automated route matching for drivers and passengers with similar itineraries and schedules, which alleviate traffic congestion and reduce travel costs. To protect the location privacy in the ride-sharing process, a variety of privacy-preserving matching schemes were proposed. However, the existing matching schemes incur high computational overhead, and cannot resist the attack of malicious adversaries. In this paper, we proposed a privacy-preserving matching scheme based on threshold private set intersection, which can protect the location privacy in the malicious model. To reduce computational overhead, we designed a new efficient method for the cardinality test of the route intersection size, based on symmetric-key primitives and a constant number of public-key operations. In addition, our privacy-preserving matching scheme uses geographic indistinguishability to generate virtual points as input sets of privacy-preserving matching scheme, and the Internet of Vehicles (IoV) uses the output of privacy-preserving matching scheme to match the appropriate vehicle for the user. The performance evaluation shows that the proposed scheme is efficient for IoV-enabled ride-sharing services. © 1967-2012 IEEE.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

An Adaptive Fuzzy SIR Model for Real-Time Malware Spread Prediction in Industrial Internet of Things Networks

引用

IEEE Internet of Things Journal 2025年第13期12卷 22875-22888页

作者： Zheng, Yan Na, Zhenyu Ji, Weidong Lu, Yang Harbin Normal University College of Computer and Information Engineering Harbin150010 China Northeast Forestry University College of Computer and Control Engineering Harbin150010 China

The Industrial Internet of Things (IIoT) networks serve as the foundational infrastructure for real-time communication and data exchange in smart manufacturing. Predicting the spread of malware within IIoT networks is particularly challenging due to uncertainties in infection and recovery rates, which are influenced by dynamic network conditions and device heterogeneity. In this paper, we propose an adaptive fuzzy SIR model that incorporates fuzzy logic and gradient descent optimization to address these uncertainties. Specifically, we integrate fuzzy logic with gradient descent, which introduces an adaptive mechanism to handle uncertain infection and recovery rates in real-time. This synergy ensures robust parameter tuning under fluctuating network states, significantly improving malware spread prediction. The proposed model dynamically adjusts infection and recovery rates using fuzzy differential equations and real-time data adaptation, enhancing prediction accuracy and resilience to network fluctuations. Experimental results demonstrate the model's advantages in improving predictive accuracy, convergence speed, and adaptability, making it a robust solution for securing IIoT networks in smart manufacturing. © 2014 IEEE.

关键词： Smart manufacturing

来源：评论

学校读者我要写书评

暂无评论

IG-DP: Graph Neural Network-based Default Prediction Method for Imbalanced Financial Datasets

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer Science 2025年第4期52卷 965-973页

作者： Wang, Changhai Ren, Jiaxi Huang, Min Liu, Qing Xu, Yaoli Software Engineering College Zhengzhou University of Light Industry Zhengzhou450000 China Software Engineering College Zhengzhou University of Light Industry Zhengzhou450000 China Software Engineering College Zhengzhou University of Light Industry Zhengzhou450000 China Electronics and Information Engineering West Anhui University Anhui Lu’an237012 China Software Engineering College Zhengzhou University of Light Industry Zhengzhou450000 China

Predicting financial defaults plays a vital role in reducing financial risks for credit businesses. A prominent trend in this area is the incorporation of borrowers’ social profiles into predictive models using graph neural networks. However, the significant imbalance between default and normal users in labeled financial datasets poses challenges for training effective prediction models, often leading to overfitting. A financial default prediction model named IG-DP is proposed in this paper, which is designed to use unlabeled background nodes to enhance prediction performance. First, imbalanced labeled samples are used to train an initial graph neural network classifier, and the node embeddings of the unlabeled samples can be obtained with the initial model. Then, background nodes proximal to the labeled samples are selected for pseudo labeling with the similarity selection module based on the embedding vector of the node. Finally, the pseudo-labeled samples are added to the training set to retrain the prediction model. The DGraphFin dataset is used for experimental evaluation, and the AUC−ROC and F1−measure are chosen as evaluation metrics. The experimental results demonstrate that IG-DP significantly outperforms other methods. Meanwhile, ablation experiments confirm the effectiveness of both the similarity selection module and the retraining loss calculation method for fusion confidence. © (2025.0), (International Association of Engineers). All Rights Reserved.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Stability Prediction in Smart Grid Using PSO Optimized XGBoost Algorithm with Dynamic Inertia Weight Updation

引用

computer Modeling in engineering & Sciences 2025年第1期142卷 909-931页

作者： Adel Binbusayyis Mohemmed Sha Department of Software Engineering College of Computer Engineering and Sciences in Al-Kharj Prince Sattam Bin Abdulaziz UniversityAl-Kharj11942Saudi Arabia

Prediction of stability in SG(Smart Grid)is essential in maintaining consistency and reliability of power supply in grid *** the fluctuations in power generation and consumption patterns of smart cities assists in effectively managing continuous power supply in the *** also possesses a better impact on averting overloading and permitting effective energy *** though many traditional techniques have predicted the consumption rate for preserving stability,enhancement is required in prediction measures with minimized *** overcome the complications in existing studies,this paper intends to predict stability from the smart grid stability prediction dataset using machine learning *** accomplish this,pre-processing is performed initially to handle missing values since it develops biased models when missing values are mishandled and performs feature scaling to normalize independent data ***,the pre-processed data are taken for training and *** that,the regression process is performed using Modified PSO(Particle Swarm Optimization)optimized XGBoost Technique with dynamic inertia weight update,which analyses variables like gamma(G),reaction time(tau1–tau4),and power balance(p1–p4)for providing effective future stability in *** PSO attains optimal solution by adjusting position through dynamic inertial weights,it is integrated with XGBoost due to its scalability and faster computational speed *** hyperparameters of XGBoost are fine-tuned in the training process for achieving promising outcomes on *** results are measured through evaluation metrics such as MSE(Mean Square Error)of 0.011312781,MAE(Mean Absolute Error)of 0.008596322,and RMSE(Root Mean Square Error)of 0.010636156 and MAPE(Mean Absolute Percentage Error)value of 0.0052 which determine the efficacy of the system.

关键词： Smart Grid machine learning particle swarm optimization XGBoost dynamic inertia weight update

来源：评论

学校读者我要写书评

暂无评论

A graph neural network technique for the prediction of cerebral stroke using an unbalanced medical dataset

引用

Multimedia Tools and Applications 2025年 1-37页

作者： Sha, Mohemmed Department of Software Engineering College of Computer Engineering and Sciences Prince Sattam Bin Abdulaziz University Al-Kharj Saudi Arabia

Cerebral stroke is a major health problem, and if not recognized and treated immediately, it can result in considerable morbidity and fatality. Predicting the possibility of a stroke can help with intervention, resulting in improved patient outcomes. However, constructing reliable stroke prediction algorithms can be difficult, especially when working with unbalanced medical data. Timely detection of the various warning signs of a stroke can significantly reduce its severity. To deal with unbalanced data, this research proposed the hybridization of two techniques Noise Reduction and oversampling technique (Noise Reduction-Border Line- Synthetic Minority Oversampling Technique (NR-BL-SMOTE) which only oversamples or strengthens the borderline minority class. It aims at eliminating noise from minority groups while preventing the making of noisy samples and BL-SMOTE is used to create synthetic data to balance the dataset. After that, feature extraction is done to extract features from the cerebral stroke prediction dataset using fine-tuned VGG19 architecture to improve the performance of the proposed model. The fine-tuned model uses an already trained network and re-trains the part of that by the new data set. Finally, classification is done using a graph neural network (GNN) to classify the prediction of normal or cerebral stroke (ischemic and hemorrhagic). Further, to optimize the accuracy of the classification performance;Improved Golden Jackal Optimization (IGJO) is done by repeating their iterations on the GNN hidden layer until we get the final optimal value. The proposed model will be simulated using the PYTHON tool. Then in the simulation results, the proposed approach achieves more powerful classification accuracy than established technologies, demonstrating its efficiency in stroke prediction and classification. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.

关键词： Graph neural networks

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

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

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：