检索结果-内蒙古大学图书馆

Multilingual Hate Speech Detection: Innovations in Optimized Deep Learning for English and Arabic Hate Speech Detection

引用

SN computer science 2025年第3期6卷 1-16页

作者： AL-Sukhani, Hassan Bsoul, Qusay Elhawary, Abdelrahman H. Nasr, Ziad M. Mansour, Ahmed E. Batyha, Radwan M. Alqadi, Basma S. Alqurni, Jehad Saad Alfagham, Hayat Madbouly, Magda M. Cybersecurity Department Faculty of Science and Information Technology Jadara University Irbid Jordan Cybersecurity Department College of Computer Sciences and Informatics Amman Arab University Amman Jordan Faculty of Engineering Ain Shams University Cairo Egypt Faculty of Computer Science Misr International University Cairo Egypt Faculty of Information Technology Applied Science Private University Amman 11931 Jordan Computer Science Department College of Computer and Information Sciences Al Imam Mohammad Ibn Saud Islamic University Riyadh Saudi Arabia Department of Educational Technologies College of Education Imam Abdulrahman Bin Faisal University P.O. Box 1982 Dammam Saudi Arabia Department of MIS College of Applied Studies and Community Service Imam Abdulrahman Bin Faisal University P.O. Box 1982 Dammam Saudi Arabia Institute of Graduate Studies & Research Alexandria University Alexandria Egypt Faculty of Computer and Data Science Alexandria University Alexandria Egypt

This paper presents the development of a multilingual hate speech detection model that effectively processes and classifies content in both Arabic and English. The study leverages both traditional machine learning models, such as K-Nearest Neighbors (KNN), Naive Bayes, and Support Vector Machines (SVM), as well as advanced deep learning models, specifically Bi-directional Long Short-Term Memory (Bi-LSTM) networks. A key challenge addressed is the classification of mixed-language content, which is common on social media platforms in the MENA region. To enhance detection performance, preprocessing techniques were applied to the text data, and the Synthetic Minority Over-sampling Technique (SMOTE) was used to balance the dataset. The results show that the Bi-LSTM model outperformed traditional machine learning approaches, particularly in identifying hate speech across multiple languages. The proposed model demonstrates superior accuracy and robustness in handling mixed-language input, providing a more effective solution for real-world hate speech detection tasks. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Class imbalance Deep learning Hate speech detection Machine learning Mixed Arabic and English training Multilingual NLP Natural language processing (NLP) Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Guest Editorial: Special Issue of Selected Extended Papers From the IEEE International Conference on Flexible, Printable Sensors and Systems (FLEPS) 2024

IEEE Journal on Flexible Electronics

引用

IEEE Journal on Flexible Electronics 2025年第1期4卷 2-3页

作者： Mantysalo, Matti Marrocco, Gaetano Agarwale, Shweta Grau, Gerd Gao, Wei Tampere University Faculty of Information Technology and Communication Sciences Tampere33720 Finland University of Roma Tor Vergata Dipartimento di Ingegneria Civile ed ed Ingegneria Informatica Rome00133 Italy Aarhus University Department of Electrical and Computer Engineering Aarhus8200 Denmark York University Department of Electrical Engineering and Computer Science TorontoONM3J 1P3 Canada Caltech Department of Engineering and Applied Science PasadenaCA91125 United States

来源：评论

学校读者我要写书评

暂无评论

Deciphering Stem Cell Fate with an Integrative Multi-Omics Examination of Microenvironmental Dynamics

引用

SN computer science 2025年第1期6卷 1-14页

作者： Shanthanam, Radha Rammohan Selvam, Janani Vajravelu, Ashok Pradeep, T. Post Doctoral Fellowship (PDF) programme in Computer Science Engineering Lincoln University College Petaling Jaya Malaysia Faculty of Engineering Lincoln University College Petaling Jaya Malaysia Faculty of Electrical and Electronic Engineering Universiti Tun Hussein Onn Malaysia Parit Raja Malaysia Kongu Engineering College Tamilnadu Erode India

The evidence-based literature on healthcare is akin to a vast universe, continually expanding with new discoveries and insights. In this dynamic landscape, the integration oftools and multi-omics approaches a beacon of hope, illuminating the intricate pathways that govern stem cell fate within the microenvironment. Still, figuring out how stem cell fate is determined is complicated and calls for a multifaceted understanding that cuts beyond traditional discipline lines. This review navigates through the evolutionary trajectory of AI integration in medicine, tracing its transformative impact on hematology and beyond. It explores the burgeoning field of AI applications in hematopoietic cell transplantation (HCT), from diagnosis to prognosis, heralding a new era of precision medicine. Through the lens of integrative multi-omics methodologies, the review delves into the molecular signatures intricately woven within the stem cell niche, deciphering the regulatory networks orchestrating stem cell fate decisions. Emerging technologies like single-cell multi-omics profiling and spatial transcriptomics offer newfound vistas to explore cellular heterogeneity and spatial organization within the stem cell niche, paving the way for transformative advances in regenerative medicine. Harnessing the power of AI and multi-omics methodologies, this review embarks on a quest to dissect stem cell fate with unparalleled precision, ushering in a new era of personalized therapeutics and regenerative interventions. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Cellular heterogeneity Hematology Molecular signatures Precision medicine Stem cell niche

来源：评论

学校读者我要写书评

暂无评论

Automated Age and Gender Recognition in Networking Sites Using Variable Optimized CNN Model 5th

Automated Age and Gender Recognition in Networking Sites Usi...

引用

5th Doctoral Symposium on Computational Intelligence, DoSCI 2024

作者： Dutta, Supratik Chakraborty, Adrita Biswal, Ashish Kumar Tripathy, Hrudaya Kumar Obaid, Ahmed J. Alkhafaji, Mohammed Ayad Kalinga Institute of Industrial Technology Deemed to be University Odisha Bhubaneswar India Faculty of Computer Science and Mathematics University of Kufa Najaf Iraq College of Technical Engineering National University of Science and Technology Dhi Qar Nasiriyah64001 Iraq

ISBN: (纸本)9789819771776

Due to the rising number of users especially in platforms like Facebook, YouTube, Instagram, and so on, automated classifying of age and gender detection has a lot of importance in this era. To implement this technology, the model is designed by detecting the age and gender of a person in separate entities to avoid confusion. Here deep learning algorithm convolutional neural network (CNN) has been used to address this problem. The OpenCV algorithm has also been used with a pre-trained Caffe model for age and gender detection. This model has the ability to identify numerous faces in a single image and forecast the gender and age of each face. The research paper showcases the rationale behind the selection of Caffe as the framework model over alternatives such as HAAR Cascade. Additionally, it presents empirical evidence indicating that the integration of OpenCV with Caffe Model yields superior outcomes compared to utilizing TensorFlow in conjunction with other models. From the 100 sample images that were taken for evaluation, an overall accuracy of 87% was achieved. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Forgery-Aware Adaptive Learning With Vision Transformer for Generalized Face Forgery Detection

引用

IEEE Transactions on Circuits and Systems for Video Technology 2025年第5期35卷 4116-4129页

作者： Luo, Anwei Cai, Rizhao Kong, Chenqi Ju, Yakun Kang, Xiangui Huang, Jiwu Kot, Alex C. Sun Yat-sen University School of Computer Science and Engineering Guangzhou510006 China Jiangxi University of Finance and Economics School of Computing and Artificial Intelligence Nanchang330013 China Laboratory School of Electrical and Electronic Engineering Jurong West 639798 Singapore Sun Yat-sen University Guangdong Key Laboratory of Information Security School of Computer Science and Engineering Guangzhou510006 China Shenzhen MSU-BIT University Guangdong Laboratory of Machine Perception and Intelligent Computing Faculty of Engineering Shenzhen518116 China China-Singapore International Joint Research Institute Guangzhou China

With the rapid progress of generative models, the current challenge in face forgery detection is how to effectively detect realistic manipulated faces from different unseen domains. Though previous studies show that pre-trained Vision Transformer (ViT) based models can achieve some promising results after fully fine-tuning on the Deepfake dataset, their generalization performances are still unsatisfactory. To this end, we present a Forgery-aware Adaptive Vision Transformer (FA-ViT) under the adaptive learning paradigm for generalized face forgery detection, where the parameters in the pre-trained ViT are kept fixed while the designed adaptive modules are optimized to capture forgery features. Specifically, a global adaptive module is designed to model long-range interactions among input tokens, which takes advantage of self-attention mechanism to mine global forgery clues. To further explore essential local forgery clues, a local adaptive module is proposed to expose local inconsistencies by enhancing the local contextual association. In addition, we introduce a fine-grained adaptive learning module that emphasizes the common compact representation of genuine faces through relationship learning in fine-grained pairs, driving these proposed adaptive modules to be aware of fine-grained forgery-aware information. Extensive experiments demonstrate that our FA-ViT achieves state-of-the-arts results in the cross-dataset evaluation, and enhances the robustness against unseen perturbations. Particularly, FA-ViT achieves 93.83% and 78.32% AUC scores on Celeb-DF and DFDC datasets in the cross-dataset evaluation. © 1991-2012 IEEE.

关键词： Forgery Faces Adaptation models Transformers Feature extraction Adaptive learning computer vision Face recognition Deepfakes Visualization

来源：评论

学校读者我要写书评

暂无评论

Ensemble Learning and Event-Based Feature Mining for Li-Ion Battery State Prediction 22

Ensemble Learning and Event-Based Feature Mining for Li-Ion ...

引用

22nd International Learning and Technology Conference, L and T 2025

作者： Qaisar, Saeed Mian Subasi, Abdulhamit Effat University Electrical and Computer Engineering Department Jeddah21478 Saudi Arabia Institute of Biomedicine Faculty of Medicine University of Turku Turku20520 Finland College of Engineering Effat University Department of Computer Science Jeddah21478 Saudi Arabia

ISBN: (纸本)9798331532215

Contemporary power solutions are revolutionizing the existing trend in residential and transportation services. Mobile realizations, in particular, are advancing based on innovative mobile power supply solutions. Batteries are essential to several key applications, such as electric vehicles (EVs), satellites, and mobile phones. Additionally, batteries are elementary parts of the energy storage for distributed renewable microgrids. The are just a few applications based on the batteries-based power supply. Li-ion batteries are frequently used because of their exceptional qualities, which include endurance, high power density, and compact size. Besides these notable benefits the Li-Ion batteries are expensive. To effectively compensate for this high initial cost, the "battery management systems"(BMSs) are used. These solutions are effective in maximizing the Li-Ion battery life and also guarantee a safer operation. Based on the advantages of modular design, the contemporary Li-Ion batteries can be comprised of thousands of individual cells. It results in complex and sophisticated BMSs and can render a notable overhead in terms of power consumption. The focus of this research is on enhancing the existing Li-Ion BMSs by incorporating novel event-based feature mining and machine/ensemble learning techniques. The event-based feature extractor mines pertinent information and promises significant real-time compression. In order to predict the capacity of Li-Ion cells, the produced feature set is then processed utilizing robust machine/ensemble learning algorithms, and the performance of considered regressors is compared. A real Li-Ion battery dataset is used to demonstrate the applicability. For the case of a bagging-based ensemble regressor, the method obtained a correlation coefficient of 0.9996. © 2025 IEEE.

关键词： Lithium-ion batteries

来源：评论

学校读者我要写书评

暂无评论

Multi-Scale Regularization and Bounding Box Refinement in Semi-Supervised Object Detection Framework 3

Multi-Scale Regularization and Bounding Box Refinement in Se...

引用

3rd International Conference on Integrated Circuits and Communication Systems, ICICACS 2025

作者： Raman, Ramakrishnan Rajan, Karthick Tlajiya, Falguni Paliwal, Manisha Pune India Chennai Institute of Technology Department of Computer Science and Engineering Chennai India Parul University Faculty of Engineering and Technology Department of Electrical Engineering Vadodara India Pune India

ISBN: (纸本)9798331508456

Semi-supervised object detection (SSOD) has emerged as a critical approach to bridge the gap between the extensive labeled data requirements of supervised methods and the abundance of unlabeled data available. However, existing SSOD techniques often suffer from two significant limitations: inconsistent performance across scales and inaccuracies in object localization. To address these challenges, this paper introduces a novel framework that integrates Multi-Scale Regularization (MSR) and Bounding Box Refinement (BBR). The proposed MSR mechanism enforces consistency in predictions across different feature map scales, ensuring robust pseudo-labeling and mitigating scale-related variations. Simultaneously, BBR employs a two-stage optimization process to enhance the precision of bounding box predictions, correcting spatial errors in initial detections. The combined framework leverages both labeled and unlabeled datasets effectively, resulting in improved detection performance. Extensive experiments conducted on benchmark datasets such as COCO and Pascal VOC demonstrate the efficacy of the proposed approach, showing significant improvements in mean Average Precision (mAP) compared to existing state-of-the-art SSOD methods. Ablation studies further highlight the individual contributions of MSR and BBR to the overall performance gains. This study establishes a new benchmark in SSOD, paving the way for more accurate and efficient object detection systems that rely on minimal labeled data while leveraging abundant unlabeled data. © 2025 IEEE.

关键词： Semi-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Evaluating the Efficiency of Dual-Zone Registration System with Timer Mechanism in Wireless Networks 3

Evaluating the Efficiency of Dual-Zone Registration System w...

引用

3rd International Conference on Integrated Circuits and Communication Systems, ICICACS 2025

作者： Raman, Ramakrishnan Vinoth, T. Barve, Amit Tiwari, Vinita Pune India Chennai Institute of Technology Department of Mechatronics Engineering Kundrathur Chennai India Paruluniversity Faculty of Engineering and Technology Department of Computer Science and Engineering Vadodara India Pune India

ISBN: (纸本)9798331508456

The increasing demand for efficient communication in wireless networks necessitates optimized registration systems that ensure smooth connectivity and scalability. This paper introduces a novel Dual-Zone Registration System with a Dynamic Timer Mechanism designed to enhance registration efficiency in wireless networks. The system classifies devices into two zones: a primary zone for immediate registration and a secondary zone for devices requiring delayed registration. Leveraging a dynamic timer mechanism, the system regulates registration attempts to reduce congestion and optimize resource allocation. Simulations reveal that the proposed system significantly improves key performance metrics, including registration success rate, average registration time, packet loss, and network load. The dual-zone system achieved a 95% registration success rate, reducing packet loss to 3% compared to 12% in centralized systems and 8% in simple timer-based systems. Average registration time decreased to 6.3 seconds, outperforming centralized (12.5 seconds) and simple timer-based systems (9.8 seconds). Additionally, the system maintained a lower network load of 45 requests per second, showcasing its scalability and efficiency. By dynamically adjusting registration delays based on real-time network conditions, the proposed system ensures efficient resource utilization and balanced load distribution, making it a promising solution for large-scale wireless networks. These findings validate the effectiveness of the dual-zone approach in improving the scalability, reliability, and overall performance of modern wireless networks. © 2025 IEEE.

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

Development and Investigation of Vision System for a Small-Sized Mobile Humanoid Robot in a Smart Environment

引用

International Journal of Crowd science 2025年第1期9卷 29-43页

作者： Abu-Jassar, Amer Tahseen Attar, Hani Amer, Ayman Lyashenko, Vyacheslav Yevsieiev, Vladyslav Solyman, Ahmed Department of Computer Science College of Information Technology Amman Arab University Amman11937 Jordan Faculty of Engineering Zarqa University Zarqa2000 Jordan College of Engineering University of Business and Technology Jeddah21448 Saudi Arabia Department of Media Systems and Technology Kharkiv National University of Radio Electronics Kharkiv61166 Ukraine Department of Electrical and Electronics Engineering Faculty of Engineering and Architecture Nişantaşı University Istanbul34398 Turkey

The conducted research aims to develop a computer vision system for a small-sized mobile humanoid robot. The decentralization of the servomotor control and the computer vision systems is investigated based on the hardware solution point of view, moreover, the required software level to achieve an efficient matched design is obtained. A computer vision system using the upgraded tiny-You Only Look Once (YOLO) network model is developed to allow recognizing and identifying objects and making decisions on interacting with them, which is recommended for crowd environment. During the research, a concept of a computer vision system was developed, which describes the interaction between the main elements, on the basis of which hardware modules were selected to implement the task. A structure of information interaction between hardware modules is proposed, and a connection scheme is developed, on the basis of which a model of a computer vision system is assembled for research, with the required algorithmic and software for solving the problem. To ensure the high speed of the computer vision system based on the ESP32-CAM module, the neural network was improved by replacing the Visual Geometry Group 16 (VGG-16) network as the base network for extracting the functions of the Single Shot Detector (SSD) network model with the tiny-YOLO lightweight network model, which made it possible to preserve the multidimensional structure of the network model feature graph, resulting in increasing the detection accuracy, while significantly reducing the amount of calculations generated by the network operation, thereby significantly increasing the detection speed, due to a limited set of objects. Finally, a number of experiments were carried out, both in static and dynamic environments, which showed a high accuracy of identifications. © The author(s) 2025.

关键词： Anthropomorphic robots

来源：评论

学校读者我要写书评

暂无评论

Weak and Extremely Low Frequency Magnetic Fields Alter the Growth Rates of HT-1080 Fibrosarcoma

Weak and Extremely Low Frequency Magnetic Fields Alter the G...

引用

2025 United States National Committee of URSI National Radio science Meeting, USNC-URSI NRSM 2025

作者： Dang, Nhat Keller, Jason Bajtos, Marek Gurhan, Hakki Barnes, Frank Department of Electrical Computer and Energy Engineering College of Engineering and Applied Science University of Colorado Boulder BoulderCO United States Department of Electromagnetics and Biomedical Engineering Faculty of Electrical Engineering and Information Technology University of Zilina Zilina Slovakia

ISBN: (纸本)9781946815200

This study presents experimental results showing that weak, extremely low frequency electromagnetic (EM) fields influence the growth rates of HT-1080 fibrosarcoma cells in vitro and offers a theoretical explanation for these effects. The results demonstrate that weak magnetic fields both increase and decrease the growth rates of fibrosarcoma cells by altering the mitochondrial superoxide (O2-) and calcium ions (Ca2+) concentrations in living cells. Over a 4-day exposure period, fibrosarcoma growth rates were modulated in the presence of low-level alternating current (AC) magnetic fields at a 10 μT amplitude at 13.5 Hz to 30 Hz frequency, superimposed on a 45 μT static background field. Effects due to changes in amplitude from 10 to 30 μT at selected frequencies are also investigated with the same superimposed static field. © 2025 USNC-URSI.

关键词： Electromagnetic field effects

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：