检索结果-内蒙古大学图书馆

Author profiling from Romanized Urdu text using transfer learning models

Neural Computing and Applications 2025年第6期37卷 4455-4470页

作者： Ali, Abid khan, Muhammad Sohail Khan, Muhammad Amin Khan, Sajid Ullah Khan, Faheem Department of Computer Software Engineering University of Engineering & Technology Mardan KPK Mardan23200 Pakistan Islamabad H-11/4 Islamabad44000 Pakistan Department of Information Systems College of Computer Engineering and Sciences Prince Sattam Bin Abdulaziz University Al-Kharj Saudi Arabia Department of Computer Engineering Gachon University Seongnam-Si Korea Republic of

This research concentrates on author profiling using transfer learning models for classifying age and gender. The investigation encompassed a diverse set of transfer learning techniques, including Roberta, BERT, ALBERT, Distil BERT, Distil Roberta, ELECTRA, and XLNet. Through meticulous evaluation using metrics such as the Matthews Correlation Coefficient, Accuracy, Precision, Recall, and F1 Score, the study examined the efficacy of these models. The curated dataset was divided for gender and age tasks, resulting in robust gender prediction with the XLNet model and age prediction with the BERT model. Notably, the XLNet model achieved the highest MCC (0.7946), Accuracy (0.8957), Precision (0.8992), Recall (0.8957), and F1 Score (0.8958) values in gender classification, while the BERT model excelled in age prediction with an MCC of (0.7338), Accuracy of (0.8220), Precision of (0.8324), Recall of (0.8220), and F1 Score of (0.8243). Visualized outcomes provide valuable insights into the model’s performance nuances, paving the way for their practical implementation. This research offers novel contributions to author profiling tasks, bridging the gap between theory and real-world applications. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Research On Fatigue Driving Detection Based On Deep Learning

引用

engineering Letters 2025年第2期33卷 348-356页

作者： Zhou, Huinan Dai, Hong University of Science and Technology Liaoning Liaoning AnshanCO 114051 China School of Computer Science and Software Engineering University of Science and Technology Liaoning Liaoning AnshanCO 114051 China

Fatigue driving is one of the main causes of traffic accidents. Effective fatigue driving detection technology can reduce traffic accidents caused by fatigue driving. Traditional fatigue driving detection methods usually use a two-stage detection method, which has the problems of low efficiency and poor accuracy. To solve these problems, this paper designs a single-stage fatigue driving detection model based on separable convolution Road Safety Net (RSNet), which extracts the features in the driver's image or video and classifies these features to determine whether the driver is in an abnormal driving state. RSNet borrows the structure of Swin Transformer and adopts a hierarchical design, which makes the model better able to extract features and classify. Each level is composed of a GDS Layer and multiple RS Blocks. To solve the problem of low efficiency of traditional methods, we use separable convolution in RS Block to replace the traditional convolutional neural network, which reduces the complexity of the model and effectively improves the computational efficiency of the model. Experimental results show that the Top-1 and Top-5 Accuracy of RSNet on the HMDB51 dataset reaches 85 % and 97.35 %, respectively, and the delay is 5.97 ms. Compared with ResNet, EfficientNet, and Swin Transformer, RSNet has higher accuracy and efficiency. In addition, this paper takes the YOLACT algorithm as an example, selects different backbone networks for object detection, and compares the indicators of mean Average Precision (mAP) and Frames Per Second (FPS). Experiments show that RSNet performs better when used as the backbone network for object detection, with mAP reaching 32.4 % and FPS of 39.6 ms. It indicates that RSNet performs well in action recognition tasks. © 2025, International Association of Engineers. All rights reserved.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Highly Robust and Diverse Coverless Image Steganography Against Passive and Active Steganalysis

引用

IEEE Transactions on Dependable and Secure Computing 2025年第3期22卷 2771-2787页

作者： Guo, Bobiao Ping, Ping Xu, Feng Hohai University College of Computer Science and Software Engineering Nanjing210098 China Hohai University Key Laboratory of Water Big Data Technology of Ministry of Water Resources Nanjing210098 China Jiangsu Ocean University School of Computer Engineering Lianyungang222005 China

To avoid the pixel modification traces left by steganography from being detected by passive steganalysis, and to prevent the hidden data from being destroyed by active steganalysis attacks, Coverless Image Steganography (CIS) that does not modify pixels has attracted widespread attention. However, most existing CIS methods are limited in their maximum capacity due to insufficient diversity in their hash sequences. In addition, these methods struggle to maintain high robustness against both geometric and non-geometric attacks simultaneously. To address these two issues, a new coverless image steganography method is proposed to enhance CIS methods’ applicability, security, and robustness in highly insecure networks. During the hiding process, hash sequences are generated by a SHA-256 algorithm that integrates inter-block and inter-channel fusion, providing higher diversity than other CIS methods. Consequently, the proposed CIS method achieves higher capacity on publicly available datasets. During the extraction process, an evaluation metric that combines visual and histogram similarity is designed to improve the accuracy of inverse image retrieval. The experimental results demonstrate that the proposed CIS method achieves capacity increases of 26.73% and 38.34% over other CIS methods on the VOC and COCO datasets, respectively. Moreover, this method exhibits nearly 100% robustness against common active steganalysis. © 2004-2012 IEEE.

关键词： Steganography Robustness Feature Extraction Data Mining Image Retrieval Visualization Accuracy software Indexes Histograms Coverless Image Steganography CIS Hash Generation Algorithm Inverse Image Retrieval Steganalysis High Robustness Robust Image Image Coding Hidden Information Retrieval Accuracy Visual Similarity COCO Dataset Convolutional Neural Network Computational Complexity Robust Method Image Dataset Bit Error Rate Types Of Attacks Auxiliary Information Scale Invariant Feature Transform Index Database Histogram Bins Length Information Cover Image Military Intelligence Secret Information Center Crop Secure Transmission Salt And Pepper Noise XOR Operation Number Of Different Sequences JPEG Compression Generative Adversarial Networks Pixel Position Common Attacks

来源：评论

学校读者我要写书评

暂无评论

An Efficient Dialogue Policy Agent with Model-Based Causal Reinforcement Learning 31

An Efficient Dialogue Policy Agent with Model-Based Causal R...

引用

31st International Conference on Computational Linguistics, COLING 2025

作者： Xu, Kai Wang, Zhenyu Zhao, Yangyang Fang, Bopeng School of Software Engineering South China University of Technology Guangdong China Department of Computer and Communication Engineering Changsha University of Science and Technology Changsha China School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China

ISBN: (纸本)9798891761964

Dialogue policy trains an agent to select dialogue actions frequently implemented via deep reinforcement learning (DRL). The model-based reinforcement methods built a world model to generate simulated data to alleviate the sample inefficiency. However, traditional world model methods merely consider one-step dialogues, leading to an inaccurate environmental simulation. Furthermore, different users may have different intention preferences, while most existing studies lack consideration of the intention-preferences causal relationship. This paper proposes a novel framework for dialogue policy learning named MCA, implemented through model-based reinforcement learning with automatically constructed causal chains. The MCA model utilizes an autoregressive Transformer to model dialogue trajectories, enabling a more accurate simulation of the environment. Additionally, it constructs a causal chains module that outputs latent preference distributions for intention-action pairs, thereby elucidating the relationship between user intentions and agent actions. The experimental results show that MCA can achieve state-of-the-art performances on three dialogue datasets over the compared dialogue agents, highlighting its effectiveness and robustness. © 2025 Association for Computational Linguistics.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Training Latency Minimization for Model-Splitting Allowed Federated Edge Learning

引用

IEEE Transactions on Network Science and engineering 2025年第3期12卷 2081-2092页

作者： Wen, Yao Zhang, Guopeng Wang, Kezhi Yang, Kun China University of Mining and Technology School of Computer Science and Technology Xuzhou221116 China Brunel University London Department of Computer Science MiddlesexUB8 3PH United Kingdom Nanjing University State Key Laboratory of Novel Software Technology Nanjing210008 China School of Intelligent Software and Engineering Nanjing University Suzhou Campus Suzhou215163 China

To alleviate the shortage of computing power faced by clients in training deep neural networks (DNNs) using federated learning (FL), we leverage the edge computing and split learning to propose a model-splitting allowed FL (SFL) framework, with the aim to minimize the training latency without loss of test accuracy. Under the synchronized global update setting, the latency to complete a round of global training is determined by the maximum latency for the clients to complete a local training session. Therefore, the training latency minimization problem (TLMP) is modelled as a minimizing-maximum problem. To solve this mixed integer nonlinear programming problem, we first propose a regression method to fit the quantitative-relationship between the cut-layer and other parameters of an AI-model, and thus, transform the TLMP into a continuous problem. Considering that the two subproblems involved in the TLMP, namely, the cut-layer selection problem for the clients and the computing resource allocation problem for the parameter-server are relative independence, an alternate-optimization-based algorithm with polynomial time complexity is developed to obtain a high-quality solution to the TLMP. Extensive experiments are performed on a popular DNN-model EfficientNetV2 using dataset MNIST, and the results verify the validity and improved performance of the proposed SFL framework. © 2013 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Runtime Verification of Neural-Symbolic Systems 10th

Runtime Verification of Neural-Symbolic Systems

引用

10th International Symposium on Dependable software engineering: Theories, Tools and Applications, SETTA 2024

作者： Deng, Shaojun Liu, Wanwei Zhang, Miaomiao College of Computer Science and Technology National University of Defense Technology Changsha China School of Software Engineering Tongji Uniersity Shanghai China

ISBN: (纸本)9789819606016

Neural-symbolic systems (NSSs), which are typically cyber-physical systems integrated with artificial intelligence modules, have received much attention in both academic and industrial fields. However, thorough verification (such as model checking) of such systems in general leads to prohibitively high costs due to the scale of the state space, and this makes it barely accomplishable within desired time-span. Consequently, light-weight verification techniques are introduced to deal with such cases, e.g., runtime verification (RV, for short). In this paper, we investigate an RV framework for NSSs against signal temporal logic (STL) properties. To guarantee that the on-line monitoring could be accomplished in time, we utilize Euler-prediction to sample the system under scrutiny at discrete time-steps. Our framework supports both qualitative and quantitative on-line monitoring of STL specifications, and thus it terminates whenever a property is satisfied and/or violated by a finite prefix of the trajectory, and it can also provide the bounds of the quantitative satisfaction/violation w.r.t. the remaining execution. Notably, for the so-called piece-wise linear NSSs against affine STL formulas, our approach is exact. On top of it, a prototype tool is implemented, and is experimentally evaluated. The results demonstrate the feasibility of the presented runtime verification approach. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Model checking

来源：评论

学校读者我要写书评

暂无评论

Dual-encoder model for typhoon path prediction with multiscale spatiotemporal data fusion

引用

Earth Science Informatics 2025年第2期18卷 1-17页

作者： Ren, Shuxia Zhong, Ruikun Guo, Zewei Zhang, Zining School of Software Engineering Tiangong University Tianjin 300387 China School of Computer Science and Technology Tiangong University Tianjin 300387 China

Extreme weather caused by typhoons poses a severe threat to human life safety and socio-economic development, making accurate prediction of typhoon paths crucial. However, existing prediction models struggle to effectively handle and integrate heterogeneous data, overlooking deep correlations between the data, which in turn affects the accuracy of the models. This paper proposes a Dual-Encoder Spatiotemporal Fusion Model for Typhoon Path Prediction (DESF-Typhoon), designed to effectively integrate data from different time scales and extract the underlying relationships between the data, thereby improving prediction accuracy. Specifically, we designed a dual-encoder module that effectively captures complex nonlinear data structures, accounting for the intricate influences of factors such as geography and environment on path, while integrating spatial features of typhoon at different time scales. Subsequently, we introduced a feature interaction module to explore the relationships between different features, which can adaptively learn the interaction weights between them, resulting in a richer feature representation. We evaluated the model using the dataset from the China Meteorological Administration (CMA) and compared its performance with traditional prediction methods and deep learning-based approaches. The results demonstrate that the model significantly improves accuracy and robustness, making innovative contributions in data fusion and spatiotemporal feature modeling. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Dual encoder Feature interaction Spatiotemporal data Tropical cyclone prediction

来源：评论

学校读者我要写书评

暂无评论

Enhancing Security in Distributed Drone-Based Litchi Fruit Recognition and Localization Systems

引用

computers, Materials & Continua 2025年第2期82卷 1985-1999页

作者： Liang Mao Yue Li Linlin Wang Jie Li Jiajun Tan Yang Meng Cheng Xiong Guangdong-Hong Kong-Macao Greater Bay Area Artificial Intelligence Application Technology Research Institute Shenzhen Polytechnic UniversityShenzhen518055China School of Computer Science and Software Engineering University of Science and Technology LiaoningAnshan114051China

This paper introduces an advanced and efficient method for distributed drone-based fruit recognition and localization, tailored to satisfy the precision and security requirements of autonomous agricultural operations. Our method incorporates depth information to ensure precise localization and utilizes a streamlined detection network centered on the RepVGG module. This module replaces the traditional C2f module, enhancing detection performance while maintaining speed. To bolster the detection of small, distant fruits in complex settings, we integrate Selective Kernel Attention (SKAttention) and a specialized small-target detection layer. This adaptation allows the system to manage difficult conditions, such as variable lighting and obstructive foliage. To reinforce security, the tasks of recognition and localization are distributed among multiple drones, enhancing resilience against tampering and data manipulation. This distribution also optimizes resource allocation through collaborative processing. The model remains lightweight and is optimized for rapid and accurate detection, which is essential for real-time applications. Our proposed system, validated with a D435 depth camera, achieves a mean Average Precision (mAP) of 0.943 and a frame rate of 169 FPS, which represents a significant improvement over the baseline by 0.039 percentage points and 25 FPS, respectively. Additionally, the average localization error is reduced to 0.82 cm, highlighting the model’s high precision. These enhancements render our system highly effective for secure, autonomous fruit-picking operations, effectively addressing significant performance and cybersecurity challenges in agriculture. This approach establishes a foundation for reliable, efficient, and secure distributed fruit-picking applications, facilitating the advancement of autonomous systems in contemporary agricultural practices.

关键词： Objective detection deep learning machine learning

来源：评论

学校读者我要写书评

暂无评论

Blockchain-enabled IoT access control model for sharing electronic healthcare data

引用

Multimedia Tools and Applications 2025年第10期84卷 8127-8148页

作者： Ilyas, Benkhaddra Kumar, Abhishek Ali, Setitra Mohamed Lei, Hang School of Information and Software Engineering University of Electronic Science and Technology of China Chengdu610054 China Chandigarh University Gharuan Mohali Punjab140413 India School of Computer Science and Engineering University of Electronic Science and Technology of China Sichuan Chengdu611731 China

Deploying the Internet of Things (IoT) in the transfer of enormous medical data often promotes challenges with the security, confidentiality, and privacy of the user’s sensitive data. In addition, the access control for prohibiting unauthorized access has been facing challenges with single-point-of-failure, collusion, and other security issues. To address these challenges associated with IoT access control, this research proposes a Blockchain-enabled IoT access control for sharing Electronic health records (EHR). The proposed authentic access control scheme transfers the data to the authenticated user without discrepancies. The proposed technique utilizes the hyperledger composer fabric blockchain that improves the transparency and traceability of the network, and the intelligent contract keeps track of the information, which reduces the possibility of unauthorized access. The proposed access control scheme validates the request effectively and transmits the data efficiently. The effectiveness of the proposed access control scheme is evaluated by analyzing the blocks in terms of metrics such as responsiveness of 185.47 secs, 95% Genuine User Detection (GUD), 96.50% Privacy, and 3.50% Information loss, for 100 blocks and proposed access control scheme works more efficiently in transmitting the data. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Smart contract

来源：评论

学校读者我要写书评

暂无评论

ViF-SD2E: a robust weakly-supervised framework for neural decoding

引用

Neural Computing and Applications 2025年第9期37卷 6645-6661页

作者： Feng, Jingyi Luo, Yong Song, Shuang Hu, Han National Engineering Research Center for Multimedia Software School of Computer Science Wuhan University Wuhan430072 China School of Information and Electronics Beijing Institute of Technology Beijing100081 China

Neural decoding plays a vital role in the interaction between the brain and the outside world. Our task in this paper is to decode the movement track of a finger directly based on the neural data. Existing neural decoding solutions primarily perform some preprocessing operations on neural data before feeding them into existing models (such as LSTM) for decoding. However, these solutions either are prone to overfitting or cannot well exploit the spatial and temporal information. In our previous observations, there is a symmetrical phenomenon between the unsupervised decoded trajectory and the ground truth trajectory within the activity space. This precisely motivates us to propose (or derive) a robust weakly-supervised framework (or model structure), called ViF-SD2E, for neural decoding. In particular, it consists of a space-division (SD) module and an exploration–exploitation (2E) strategy, to effectively exploit both the spatial information of the outside world and the temporal information of neural activity, where the SD2E output is analogized with the weak 0/1 vision feedback (ViF) label for training. Extensive experiments demonstrate the effectiveness of our method, which can sometimes be comparable to supervised counterparts. Therefore, we redirect our attention to the information (hidden in data) ViF-SD2E conveys to us. In other words, we believe that the advantage of ViF-SD2E lies in the fact that its processing steps are objectively determined by the inherent attributes (i.e., symmetry) of the neural data, or rather, the model structure is fixed. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： Space division multiple access

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：