检索结果-内蒙古大学图书馆

Computational Intelligence Approaches and Applications (ICCIAA), International Conference on

作者： Ola Nasir Waleed Amer Ala Hamarsheh Ali Mahmoud Ali Amani Abu Zaid Department of Computer Science Faculty of Information Technology Zarqa University Zarqa Jordan Department of Cyber Secyrity Faculty of Information Technology Zarqa University Zarqa Jordan Department of computer science Faculty of Information Technology Arab American University Department of Computer Network Faculty of Engineering Technology Al-Balqa Applied University Amman Jordan

ISBN: (数字)9798331523657

ISBN: (纸本)9798331523664

Zero-Day attacks represent one of the most challenging threats in modern cybersecurity, as they exploit unknown vulnerabilities and bypass traditional detection systems. In this paper, we propose a hybrid threat detection system that integrates three key components: a Random Forest-based supervised classification model for detecting known attack patterns, an autoencoder-based anomaly detection module to uncover novel threats, and a Reinforcement Learning (RL) agent to dynamically optimize response strategies in real-time. The proposed system achieves a detection Accuracy of 97.8% for known attacks. The autoencoder, trained exclusively on normal traffic, employs a dynamically determined threshold (set at 0.8409) to flag anomalous behavior indicative of Zero-Day attacks. Furthermore, the RL agent adapts quickly to evolving threats, effectively filtering out 53.2% of benign traffic while blocking 39.7% of attack traffic and quarantining 7.1% of ambiguous cases for further inspection. Experimental results demonstrate that this integrated approach not only enhances top-level threat prevention but also significantly reduces false alerts.

关键词： Heuristic algorithms Prevention and mitigation Autoencoders Reinforcement learning Threat assessment Real-time systems computer security Anomaly detection Random forests Testing

来源：评论

学校读者我要写书评

暂无评论

Performance Evaluation of Multi-ONU Customers Access Network using Spectrum Slicing and Duplication Techniques 5

Performance Evaluation of Multi-ONU Customers Access Network...

引用

5th International Conference on Emerging Electrical Energy, Electronics and Computing Technologies, ICE4CT 2024

作者： Rahman, Juwairiyyah Abdul Mugunthan, Sivaguru A.L. Hassan, Hasliza Abu Muslim, Norliana Ab-Rahman, Mohammad Syuhaimi Department of Electrical Engineering Faculty of Engineering and Life Sciences Universiti Selangor Selangor Bestari Jaya45600 Malaysia Department of Computer and Communication Technology Faculty of Information and Communication Technology Universiti Tunku Abdul Rahman Perak Kampar31900 Malaysia Department of Electrical Electronics and Systems Engineering Faculty of Engineering and Built Environment Universiti Kebangsaan Malaysia Malaysia Electrical and Computer Engineering Program Faculty of Engineering Sohar University PC:311 Sohar Oman

This study aims to increase the number of access users by limiting the sample size to 30 users while ensuring that every Optical Network Unit (ONU) receives data from the Optical Line Terminal (OLT). The proposed solution allows for the determination of power requirements at each ONU based on bus-star topologies, facilitating an increase in user capacity. The power received by each ONU is a critical parameter that system designers must consider when planning the implementation of Fiber to the Home (FTTH), as it impacts the output data in terms of Bit Error Rate (BER) and Signal Quality Factor (QF) for each user. Various parameters were evaluated to ensure that the enhanced network improves scalability while adhering to standard specifications. These parameters included the Q-factor, BER, eye diagram, user count, and received power. The study employed slicing and duplication techniques, utilizing Arrayed Waveguide Grating (AWG) for signal duplication and Wavelength Division Multiplexing for signal slicing. The simulations were conducted using a transmitter operating at a frequency of 1491 nm with a power output of 0 dBm, utilizing the latest version of OptiSystem 20 software. The results highlighted the system's capability to support a total of 73,728 users, meeting stringent performance criteria. Specifically, the system achieved a Q-factor greater than 6, indicating high signal quality, and a bit error rate (BER) lower than 1 × 10-9, ensuring reliable data transmission. For the sampled 30 users, the average Q factor was 8.1527, and the average BER was 5.3648 × 10^-11, with an output power level of -54.278 dBm. Despite an increase in users, the quality of the data supplied to clients remained unchanged. These outcomes underscore the robustness and efficiency of the design in delivering high-capacity, error-resilient optical communication. © 2025 Institute of Physics Publishing. All rights reserved.

关键词： Q factor measurement

来源：评论

学校读者我要写书评

暂无评论

A Comparative Study of Validation Methods for Sensor-Based Human Activity Recognition Using Deep Learning Models 10

A Comparative Study of Validation Methods for Sensor-Based H...

引用

10th International Conference on Digital Arts, Media and technology and 8th ECTI Northern Section Conference on Electrical, Electronics, computer and Telecommunications engineering, ECTI DAMT and NCON 2025

作者： Mekruksavanich, Sakorn Hnoohom, Narit Phaphan, Wikanda Jitpattanakul, Anuchit University of Phayao School of Information and Communication Technology Department of Computer Engineering Phayao Thailand Mahidol University Image Information and Intelligence Laboratory Department of Computer Engineering Faculty of Engineering Nakhon Pathom Thailand King Mongkut's University of Technology North Bangkok Research Group in Statistical Learning and Inference Department of Applied Statistics Faculty of Applied Science Bangkok Thailand King Mongkut's University of Technology North Bangkok Intelligent and Nonlinear Dynamic Innovations Research Center Department of Mathematics Faculty of Applied Science Bangkok Thailand

ISBN: (纸本)9798331543273

With the rise of wearable sensors and smart devices, human activity recognition (HAR) has become a vital research area in ubiquitous computing. Although many studies report high accuracy using k-fold cross-validation, these results often do not reflect actual generalization interpretation due to subject-dependent data leakage, where models test on activities from the subjects they were trained. This study compares traditional k-fold cross-validation with leave-one-subject-out (LOSO) validation using the WISDM dataset, highlighting the importance of proper validation techniques in HAR systems. We implemented and evaluated five advanced deep learning models - convolutional neural network (CNN), long short-term memory (LSTM), bidirectional LSTM (BiLSTM), gated recurrent unit (GRU), and bidirectional GRU (BiGRU) - on windowed raw sensor data. Our experiments showed significant differences between validation methods. The BiGRU model achieved 97.91% accuracy with k-fold cross-validation and 98.02% with LOSO validation, while the CNN model achieved only 92.55% and 93.44%, respectively. These results underscore the impact of both model architecture and validation approach on performance. Our findings stress the need for subject-independent validation strategies like LOSO to develop truly generalizable HAR systems. © 2025 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Emotion Recognition Using EEG Signals in Human Brain Waves and Deep Learning Approaches 10

Emotion Recognition Using EEG Signals in Human Brain Waves a...

引用

作者： Mekruksavanich, Sakorn Hnoohom, Narit Phaphan, Wikanda Jitpattanakul, Anuchit University of Phayao School of Information and Communication Technology Department of Computer Engineering Thailand Mahidol University Image Information and Intelligence Laboratory Department of Computer Engineering Faculty of Engineering Nakhon Pathom Thailand King Mongkut's University of Technology North Bangkok Research Group in Statistical Learning and Inference Department of Applied Statistics Faculty of Applied Science Bangkok Thailand King Mongkut's University of Technology North Bangkok Intelligent and Nonlinear Dynamic Innovations Research Center Department of Mathematics Faculty of Applied Science Bangkok Thailand

ISBN: (纸本)9798331543273

Emotion recognition using EEG signals has gained growing attention in recent years due to its promising applications in human-computer interaction, affective computing, and mental health assessment. This study investigates the performance of various deep learning architectures in classifying emotional states from EEG recordings. We utilize the Emotion EEG dataset as a benchmark and evaluate five prominent deep learning models. The classification task targets three emotion categories: neu-tral, positive, and negative. Experimental results reveal notable differences in model performance. Among the tested models, BiGRU achieves the highest classification accuracy of 97.70% (±0.69%), followed closely by BiLSTM with 97.37% (±0.17%). These bidirectional architectures significantly outperform the CNN model, which attains an accuracy of only 61.59% (±1.22%). The LSTM and GRU models yield intermediate results, nearing 92% accuracy. The findings highlight the effectiveness of bidirectional recurrent neural networks in modeling the temporal characteristics of EEG signals for emotion detection. This research contributes to advancing EEG-based emotion recognition by identifying the most suitable deep learning strategies, thus supporting the development of more intelligent and emotion-aware computing systems. © 2025 IEEE.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Orthopedic Walker Fall Detection and Motion Classification Using Bidirectional Gated Recurrent Unit Neural Network 10

Orthopedic Walker Fall Detection and Motion Classification U...

引用

作者： Mekruksavanich, Sakorn Hnoohom, Narit Phaphan, Wikanda Jitpattanakul, Anuchit University of Phayao School of Information and Communication Technology Department of Computer Engineering Thailand Mahidol University Image Information and Intelligence Laboratory Department of Computer Engineering Faculty of Engineering Nakhon Pathom Thailand King Mongkut's University of Technology North Bangkok Research Group in Statistical Learning and Inference Department of Applied Statistics Faculty of Applied Science Bangkok Thailand King Mongkut's University of Technology North Bangkok Intelligent and Nonlinear Dynamic Innovations Research Center Department of Mathematics Faculty of Applied Science Bangkok Thailand

ISBN: (纸本)9798331543273

Accurate fall detection for orthopedic walker users is essential for timely medical intervention and enhancing safety for elderly and mobility-impaired individuals. This paper introduces a new method using a bidirectional gated recurrent unit (BiGRU) neural network to detect falls and classify motion patterns in walker usage. We use the Walker dataset, which includes inertial measurement unit sensor data for four motion classes: idle, motion, step, and fall. The dataset consists of 6-axis IMU readings (3-axis accelerometer and 3-axis gyroscope) from sensors mounted on orthopedic walkers, with around 620 samples per class. Our BiGRU model leverages temporal dependencies in both forward and backward directions to improve classification accuracy. We assess the model's performance in two scenarios: binary classification (fall vs. non-fall) and multi-class classification of all four motion types. The proposed architecture shows strong performance, achieving 99.79% accuracy and a 99.73% F1-score in binary classification, and 98.51% accuracy and a 98.50% F1-score in multi-class classification. It outperforms traditional deep learning models like CNN, LSTM, and GRU. The experimental results confirm the effectiveness of bidirectional processing in capturing temporal motion patterns, indicating significant potential for realtime fall detection in walker-assisted mobility. The high precision and recall values in both scenarios demonstrate the model's reliability for practical healthcare monitoring applications. © 2025 IEEE.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

An intelligent framework for sarcasm detection in Arabic tweets using deep learning with Al-Biruni earth radius optimization algorithm

引用

Alexandria engineering Journal 2025年 127卷 562-572页

作者： AL-anazi, Reema G. A. Alzaidi, Muhammad Swaileh Eltahir, Majdy M. Almukadi, Wafa Sulaiman Alajmani, Samah Hazzaa Darem, Abdulbasit A. Alshahrani, Mohammed Alhebaishi, Nawaf Department of Arabic Language and Literature College of Humanities and Social Sciences Princess Nourah bint Abdulrahman University P.O. Box 84428 Riyadh11671 Saudi Arabia Department of English Language College of Language Sciences King Saud University P. O. Box 145111 Riyadh Saudi Arabia Department of Information Systems Applied College at Mahayil King Khalid University Saudi Arabia Department of Software Engineering College of Engineering and Computer Science University of Jeddah Saudi Arabia Department of Information Technology College of Computers and Information Technology Taif University P.O. Box 11099 Taif21944 Saudi Arabia Department of Computer Science College of Science Northern Border University Arar Saudi Arabia Computer Department Applied College Najran University Najran66462 Saudi Arabia Department of Information Systems Faculty of Computing and Information Technology King Abdulaziz University Jeddah Saudi Arabia

Social media networks play a major role in expressing people's feelings, reviews, and thoughts. One of the popular linguistic patterns to express or criticize one's ideas with ridicule is sarcasm, where they have unintended and intended meanings in written text. The sarcastic text is used to reverse the polarity of the sentiment. As a result, sarcasm detection in the text positively affects the sentimental analysis (SA) tasks and improves accuracy. The sarcasm recognition of Arabic content is constrained even though Arabic is the most popular language for web content sharing, and yet still naive owing to various challenges, such as the insufficient data sources, the multiple dialects, and the morphological structure of the Arabic language. This study presents an Artificial Intelligence Driven Sarcasm Detection using Optimal Deep Learning (AIDSD-ODL) technique on Arabic tweets. In the AIDSD-ODL technique, a hyperparameter-tuned DL model is utilized to identify sarcastic or non-sarcastic Arabic tweets. At the initial stage, the AIDSD-ODL method undergoes data pre-processing to convert the input tweets into a compatible format. In the next phase, the Glove word embedding process is used in the AIDSD-ODL technique. The self-attention bidirectional long short-term memory (SA-BiLSTM) model is utilized for sarcasm detection. The Al-Biruni Earth Radius (BER) model is implemented for the hyperparameter selection process to enhance the performance of the SA-BiLSTM network. The experimental results of the AIDSD-ODL method are examined under the Arabic tweets dataset. The comparison study of the AIDSD-ODL method portrayed a superior accuracy value of 94.20 % over existing models. © 2025 The Authors

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Data-Driven Group Formation Control of Cyber-Physical Systems via Distributed Cloud Computing

IEEE Transactions on Industrial Cyber-Physical Systems

引用

IEEE Transactions on Industrial Cyber-Physical Systems 2025年 3卷 341-350页

作者： Ren, Hongru Long, Yinren Li, Hongyi Huang, Tingwen Guangdong University of Technology School of Automation Guangdong-Hong Kong Joint Laboratory for Intelligent Decision and Cooperative Control Guangdong Provincial Key Laboratory for Intelligent Decision and Cooperative Control Guangzhou510006 China Southwest University College of Electronic and Information Engineering Chongqing400715 China Shenzhen University of Advanced Technology Faculty of Computer Science and Control Engineering Shenzhen518055 China

This paper investigates the group formation control problem for cyber-physical systems (CPSs) with random communication constraints. The distributed cloud computing system is constructed to divide agents into groups and establish communication between agents. A data-driven predictive control strategy is proposed by combining networked predictive control and model-free adaptive control method. The desired group formation control performance can be achieved and the three-channel random communication constraints of CPSs are actively compensated. Thisstrategy does not require the system model and relies solely on the system's I/O data for adaptive learning. Further analyses concludes the conditions for simultaneous reach stability and group formation of the closed-loop CPSs using the data-driven predictive control strategy. The effectiveness of the proposed strategy is validated by simulation results. © 2023 IEEE.

关键词： Distributed cloud

来源：评论

学校读者我要写书评

暂无评论

Deployment of neural networks for WLAN thumbprint indoor positioning system

引用

International Journal of information technology (Singapore) 2025年 1-7页

作者： Kushwaha, Ajay Agarwal, Trapty Singh, Manpreet MP, Sunil Sharma, Amit Jain, Romil Bhuvaneswari, P. Computer Science and Engineering Rungta College of Engineering and Technology Chhatisgarh Bhilai India Maharishi School of Engineering & Technology Maharishi University of Information Technology Uttar Pradesh Lucknow India Centre of Research Impact and Outcome Chitkara University Punjab India Department of Electronics and Communication Engineering Faculty of Engineering and Technology Jain (Deemed-to-be University) Karnataka Bangalore India School of Computer Applications Lovely Professional University Punjab India Chitkara Centre for Research and Development Chitkara University Himachal Pradesh Chandigarh India Department of Electronics and Communication Engineering Vel Tech Rangarajan Dr Sagunthala R & D Institute of Science and Technology Chennai India

With the application and development of related technologies based on user location information, location-based services have become essential for daily lives. Despite this fact, Wireless Local area networks (WLAN’s) often face time complexities in streamlining and collection of data. Thus, the given paper introduces a deep-learning model based on convolutional neural network (CNN) to predict user’s indoor location from the received wireless signal strength fingerprint maps. The aim is to enhance WLAN indoor localization by streamlining the data collection process. The proposed indoor positioning model leverages a region of interest (ROI) extraction technique to simulate the wireless received signal strength intensity in WLAN’s. By incorporating this advanced technique, the model aims to improve the accuracy and efficiency of WLAN localization. To optimize and evaluate the model’s performance in real-world scenarios, experiments are conducted using simulated datasets obtained from the publicly available database such as UCI repository. The results show that the proposed model attains the highest validation accuracy (97.11%) and training accuracy (97.34%) as compared to the existing deep learning algorithms used for indoor localization. © Bharati Vidyapeeth's Institute of computer Applications and Management 2024.

关键词： CNN Deep learning Indoor localization Region of interest (ROI) Signal strength WLAN

来源：评论

学校读者我要写书评

暂无评论

AI in Cancer Research: Challenges, Applications, and Future Directions

AI in Cancer Research: Challenges, Applications, and Future ...

引用

Conference of Recent Trends and Applications of Soft Computing in engineering, RTASCE 2024

作者： Selimovic, Mirza Almisreb, Ali Abd Ismail, Nurlaila Amanzholova, Saule Faculty of Computer Science and Engineering International University of Sarajevo Sarajevo Bosnia and Herzegovina College of Engineering Universiti Teknologi MARA Selangor Shah Alam Malaysia Department of Cybersecurity International Information Technology University Almaty Kazakhstan

ISBN: (纸本)9783031828805

The era of cancer diagnosis and treatment can change dramatically through Artificial Intelligence (AI). The present paper looks into AI's transformative attributes alongside the barriers and constraints that need to be solved for clinical integration to be successful. We discuss how AI-assisted solutions can help target therapeutic regime investigation and speed up medical decision-making in different medical fields related to oncology. Next, we highlight the challenges that harnessing AI faces like data standardization, ethical matters, and the level of accuracy of the data and adaptability of AI models. In summary, we underline the necessity of international cooperation, the importance of XAI (explainable AI) for establishing trust, and the conduct of the research in a manner that involves clinical trials and real-time demonstration in patients to guarantee database safety and usefulness in cancer care. AI can change the treatment world of cancers by combatting these obstacles and moving onto new ways. Such possibilities entail more accurate diagnosis, effective therapy, and hopefully, improved outcomes for the patients in the end. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Theranostics

来源：评论

学校读者我要写书评

暂无评论

Exploring the Dark Side: A Systematic Review of Generative AI’s Role in Network Attacks and Breaches

Exploring the Dark Side: A Systematic Review of Generative A...

引用

Conference of Recent Trends and Applications of Soft Computing in engineering, RTASCE 2024

作者： Kurtović, Hamza Šabanović, Esma Almisreb, Ali Abd Saleh, Mohammed A. Ismail, Nurlaila Faculty of Computer Science and Engineering International University of Sarajevo Sarajevo Bosnia and Herzegovina Department of Cybersecurity International Information Technology University Almaty Kazakhstan College of Engineering Universiti Teknologi MARA Selangor Shah Alam Malaysia

ISBN: (纸本)9783031828805

Generative artificial intelligence has a paradoxical nature, which on the one hand brings innovative breakthroughs, and on the other hand brings with it a series of growing challenges. Processing of natural language, computer vision and generating creative content has been significantly changed and redefined, and therefore its darker background emerges through exploitation in cybercrime, which increases the complexity of protecting network systems and infrastructure. The paper explores the conflicting characteristics of generative artificial intelligence in the field of cybersecurity, focusing on the troubling issues it creates. With the advancement of generative artificial intelligence, powerful tools have been discovered to orchestrate new types of cyberattacks, and such attacks threaten not only the confidentiality and integrity of data, but also pose significant threats to business viability and national security. The research conducts a thorough analysis of the literature to explore the role of generative artificial intelligence in these attacks, highlighting its dual function as an enabler of technological progress and a facilitator of cyber risk. An in-depth examination of the vulnerabilities present in network systems and applications that are particularly vulnerable to generative attacks driven by artificial intelligence are sought to be addressed in the paper. Additionally, it assesses future defense methods and technological breakthroughs with the potential to reduce the dangers associated with generative artificial intelligence in the digital world, and aims to provide a special perspective on the dynamic interaction between generative artificial intelligence and cybersecurity issues by analyzing and studying previous research and emphasizing the need to create strong and effective solutions. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Generative adversarial 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

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：