检索结果-内蒙古大学图书馆

UPB Scientific Bulletin, Series C: Electrical engineering and computer Science 2024年第2期86卷 61-72页

作者： Dorobăț, Ilie Cristian Posea, Vlad Boncea, Radu Dept. of Computer Science National University of Science and Technology POLITEHNICA Bucharest Romania National Institute for Research and Development in Informatics-ICI Bucharest Romania

The paper introduces the Cultural Leaf Semantic Portal whose aim is to provide a consolidated user-friendly solution for exploring the 36,132 Romanian artifacts and the 1,086 cultural entities that house them. Using the eCHO framework, the data collected from official sources are subjected to a transformation process through which they are sanitized, temporal expressions are normalized, thus reducing the number of forms in which the centuries and millennia are expressed from 1,733 to only 67, and then they are transformed into semantic structures. Finally, the resulting data is accessible via both a SPARQL endpoint and a semantic portal. © 2024, Politechnica University of Bucharest. All rights reserved.

关键词： Semantic Web

来源：评论

学校读者我要写书评

暂无评论

Sensitive Content Detection in Social Networks Using Deep Learning Models and Explainability Techniques 9

Sensitive Content Detection in Social Networks Using Deep Le...

引用

9th IEEE/ACIS International Conference on Big Data, Cloud Computing, and Data Science, BCD 2024

作者： Perikos, Isidoros University of Patras Computer Engineering and Informatics Department Patras Greece University of Peloponnese Computer Technology Institute and Press 'Diophantus' Electrical and Computer Engineering Dept. Patras Greece

ISBN: (纸本)9798350394191

In the dynamic landscape of online social networks, recognizing sensitive content is essential for safeguarding user privacy, fostering inclusivity, and enhancing diversity awareness. Building on prior research, this study explores new dimensions and methodologies for detecting sensitive content. We examine the temporal evolution of sensitive content, revealing how patterns shift over time, and address cross-linguistic challenges, emphasizing cultural and contextual nuances in detection. We employ advanced machine learning techniques, including deep learning models and BERT that improve the accuracy and robustness of the detection procedure. In the experimental study, BERT transformer reported the best performance in detecting sensitive content in text. Additionally, we incorporate explainability techniques such as LIME and SHAP to provide deeper insights into the model's decision-making processes, ensuring predictions are interpretable and reliable. Our work enhances the theoretical framework of sensitive content detection in social networks and provide methods that are accurate and scalable and can facilitate the creation of user-centric interaction that prioritize privacy and user experience. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Approaches for Stress Detection: A Survey

引用

IEEE Transactions on Affective Computing 2024年第2期16卷 499-517页

作者： Kyrou, Maria Kompatsiaris, IoannisSenior Petrantonakis, Panagiotis C. Aristotle University of Thessaloniki Artificial Intelligence and Information Analysis Laboratory Dept. of Informatics GR 54124 Greece Information Technologies Institute Centre for Research and Technology Hellas GR 57001 Greece Aristotle University of Thessaloniki Dept. of Electrical and Computer Engineering GR 54124 Greece

Stress has a severe impact on individuals irrespective of age, sex, work, or background. The reliable development of stress detection techniques enhances the social, educational, physical, economic, and professional quality of life, preventing chronic stress and proposing alleviation strategies. research studies examine psychological, cognitive, behavioral, and physiological reactions to identify stress adequately. Deep Learning (DL) has received significant attention in recent years as it deals with high-dimensional, heterogeneous data and automatically learns representative features. This paper presents a survey on stress detection with recent DL approaches, leveraging data from all possible sources (physiological, speech, facial expressions, gestures, and social media content). The methodological outlines, the best results, and the main contributions of each study are discussed. We also describe publicly available datasets used by several of the presented works. Finally, we emphasize various open issues within the field of research and highlight key directions for future work © 2024 IEEE.

关键词： Social psychology

来源：评论

学校读者我要写书评

暂无评论

A Comparative Analysis of Transformer Models in Zero-Shot Text Classification 15

A Comparative Analysis of Transformer Models in Zero-Shot Te...

引用

15th International Conference on Information, Intelligence, Systems and Applications, IISA 2024

作者： Kyritsis, Konstantinos Liapis, Charalampos M. Spatiotis, Nikolaos Perikos, Isidoros Paraskevas, Michael University of Peloponnese Electrical & Computer Engineering Dept. Patras26334 Greece University of Patras Mathematics Dept. Patras26504 Greece Computer Technology Institute & Press 'Diophantus' Patras26504 Greece University of Patras Computer Engineering & Informatics Dept. Patras26504 Greece

ISBN: (纸本)9798350368833

Ahstract-Zero-shot text classification leverages pre-trained transformer models to categorize texts without the need for task-specific training. This paper analyzes a variety of transformer-based pre-trained methods on the task of zero-shot text classi-fication. Specifically, we present a deep comparative analysis of various transformer models, such as BART, DeBerta, DistiIBART, RoBERTa, and their variants, and evaluate their performance in various zero-shot text classification schemes. Furthermore, we examine the models' generalization capabilities. The findings highlight key strengths and weaknesses of each model, providing insights into their suitability for different text categorization tasks. Our research contributes to the broader understanding of transformer-based approaches and offers guidance for selecting models for zero-shot text classification. © 2024 IEEE.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

Zero-Shot Sentiment Analysis Exploring BART Models 8

Zero-Shot Sentiment Analysis Exploring BART Models

引用

8th IEEE/ACIS International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023

作者： Kyritsis, Konstantinos Perikos, Isidoros Paraskevas, Michael University of Peloponnese Computer Technology Institute and Press "diophantus" Electrical and Computer Engineering Dept. Patras Greece University of Patras Computer Engineering and Informatics Department Patras Greece

ISBN: (纸本)9798350373615

The BART model is an advanced adaptation of transformers introduced by Facebook. It has incorporated elements from both BERT and GPT transformers, enabling significant advancements in language understanding and general speech processing. Utilizing both encoder and decoder components, BART proves versatile for various tasks, including translation, text completion, automatic sentence generation, entity recognition, sentiment analysis, and more. In this study, we focus on the study of pretrained models, BART and a modified version called distilbart, in the context of Zero-Shot Text Classification. In the experimental study we dive into the Zero-Shot technique applied to various pretrained Transformers. Our analysis demonstrates that, depending on the Model we utilize, we can achieve F1 scores of up to 88%, showcasing the model's effectiveness in discerning classes for this Sentiment Analysis problem using the Zero-Shot Text Classification technique. © 2023 IEEE.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Implementing a Robust and Scalable Detect-and-Avoid Algorithm for UAV Navigation in ROS 15

Implementing a Robust and Scalable Detect-and-Avoid Algorith...

引用

15th International Conference on Information, Intelligence, Systems and Applications, IISA 2024

作者： Meimetis, Dimitrios Daramouskas, Ioannis Lappas, Vaios Perikos, Isidoros Kapoulas, Vaggelis Paraskevas, Michael Nkua Department of Aerospace Science & Technology Athens Greece Computer Technology Institute and Press 'Diophantus' Patras Greece UoP Applied Mechanics Lab Patras Greece UoP Computer Engineering and Informatics Department Patras Greece Computer Technology Institute and Press 'Diophantus' University of Peloponnese Electrical and Computer Engineering Dept. Patras Greece

ISBN: (纸本)9798350368833

Unmanned aerial vehicles (UAVs) have garnered significant attention from the research community during the last decade, due to their diverse capabilities and potential applications. One of the most critical functions that drones must execute efficiently is navigation in real-world environments. This paper presents a decentralized approach for enabling unmanned aerial vehicles (UAVs) to navigate safely in unknown environments and avoid obstacles. Leveraging the Optimal Reciprocal Collision Avoidance (ORCA) algorithm, implemented in the Robot Operating System (ROS), our method facilitates conflict detection and resolution in 2D environments. Through simulations using ROS, Gazebo, and Iris drones, we validate the effectiveness of our approach in scenarios with initial trajectory conflicts. Our work addresses the pressing need for UAVs to autonomously plan and execute safe flights, laying the groundwork for enhanced UAV capabilities in various real-world applications. The simulation results demonstrate the efficiency and robustness of our approach. © 2024 IEEE.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

Context-aware Container Orchestration in Serverless Edge Computing

Context-aware Container Orchestration in Serverless Edge Com...

引用

2024 IEEE Global Communications Conference, GLOBECOM 2024

作者： Guan, Peiyuan Chen, Chen Chen, Ziru Cai, Lin X. Hao, Xing Taherkordi, Amir University of Oslo Dept. of Informatics Oslo Norway University of Cambridge Dept. of Computer Science and Technology Cambridge United Kingdom Illinois Institute of Technology Dept. of Electrical and Computer Engineering Chicago United States Northwest Univesity School of Information Science and Technology Xi'an China

ISBN: (纸本)9798350351255

Adopting serverless computing to edge networks benefits end-users from the pay-as-you-use billing model and flexible scaling of applications. This paradigm extends the boundaries of edge computing and remarkably improves the quality of services. However, due to the heterogeneous nature of computing and bandwidth resources in edge networks, it is challenging to dynamically allocate different resources while adapting to the burstiness and high concurrency in serverless workloads. This article focuses on serverless function provisioning in edge networks to optimize end-to-end latency, where the challenge lies in jointly allocating wireless bandwidth and computing resources among heterogeneous computing nodes. To address this challenge, We devised a context-aware learning framework that adaptively orchestrates a wide spectrum of resources and jointly considers them to avoid resource fragmentation. Extensive simulation results justified that the proposed algorithm reduces over 95% of converge time while the end-to-end delay is comparable to the state of the art. © 2024 IEEE.

关键词： Quality of service

来源：评论

学校读者我要写书评

暂无评论

Resource Efficient Federated Learning for Deep Anomaly Detection in Industrial IoT applications 24

Resource Efficient Federated Learning for Deep Anomaly Detec...

引用

24th International Conference on Digital Signal Processing, DSP 2023

作者： Gkillas, Alexandros Lalos, Aris Industrial Systems Institute Athena Research Center Greece University of Patras Dept. of Computer Engineering and Informatics Greece

ISBN: (纸本)9798350339598

Anomaly data constitute a thorny problem in numerous industrial applications. In recent years, deep learning enabled anomaly detection has emerged as a critical direction, however the improved detection accuracy and reconstruction is achieved with the utilization of large neural networks with more layers and nodes, increasing their storage and computational cost. Moreover, the data collected in edge devices contain user privacy, introducing challenges that can be successfully addressed by the very recent on-device privacy-preserving distributed machine learning paradigm, known as federated learning (FL). This paradigm allows edge devices to locally train and exchange models increasing also the communication cost. Thus, in order to deal with the increased communication, processing and storage challenges introduced by FL based deep anomaly detection NN pruning is expected to have significant benefits towards both reducing the processing, storage and communication complexity but also towards avoiding the over fitting problem. With this focus, a novel compression-based optimization problem is proposed at the server-side of a FL paradigm that simultaneously fusses the local models broadcast by the edge devices and performs pruning generating a much more compressed model to be deployed at the edge devices with no accuracy loss even if the training is performed using a small subset of data instances, achieving compression rates greater than 99%. © 2023 IEEE.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Secure Sentinel Leveraging Machine Learning for Fraud Detection in Blockchain Transactions 5

Secure Sentinel Leveraging Machine Learning for Fraud Detect...

引用

5th International Conference on Data Intelligence and Cognitive informatics, ICDICI 2024

作者： Shirley, C.P. Thanga Helina, S. Berin Jeba Jingle, I. Saran, P. Absin, S.J. Karunya Institute of Technology and Sciences Dept. of Computer Science and Engineering Coimbatore India Kpr College of Arts Science and Research Department of Commerce with Computer Application Coimbatore India

ISBN: (数字)9798350389609

ISBN: (纸本)9798350389609

Blockchain technology's decentralized and transparent transaction platforms have changed a number of sectors. However, harmful activities like fraud and money laundering are also drawn to blockchain due to its distributed and unchangeable nature. The distinctive features of blockchain transactions frequently prove to be too much for conventional fraud detection techniques to handle. In order to identify fraud in blockchain transactions, this work suggests "Secure Sentinel,"a machine learning-based method based on the Isolation Forest concept. By separating anomalies in a binary tree structure, the Isolation Forest algorithm is highly effective in detecting abnormalities inside datasets. By utilizing this approach, our system is able to identify transactions on the blockchain that are suspicious or stray from the norm. The Isolation Forest model will be trained using attributes such transaction amount, frequency, source/destination addresses, and transaction time. The project's goal is to improve blockchain networks' security and integrity by offering a real-time fraud detection system. Users, miners, and regulators are among the blockchain stakeholders who can take proactive steps to reduce risks and preserve system confidence by automatically identifying potentially fraudulent transactions. We will use past blockchain data to test our model's performance and determine how well it detects different kinds of fraudulent activity in terms of accuracy, precision, recall, and efficiency. By using machine learning to protect blockchain ecosystems from new dangers, Secure Sentinel is a step in the right direction toward advancing the acceptance and sustainability of blockchain technology. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Velocity-aided Channel Estimation for Spatially Selective mmWave Massive MIMO Communications 31

Velocity-aided Channel Estimation for Spatially Selective mm...

引用

31st European Signal Processing Conference, EUSIPCO 2023

作者： Vlachos, Evangelos Mavrokefalidis, Christos Berberidis, Kostas Industrial Systems Institute ATHENA Research Center Patras Greece Dept. of Computer Engineering and Informatics University of Patras Patras Greece

ISBN: (纸本)9789464593600

This paper focuses on addressing the challenge of estimating multiple-input multiple-output (MIMO) channels for wireless communication between a ground base-station and a moving vehicle. One recently recognised model for time-varying channels incorporates spatial selectivity, which is referred to as beam squint, and is particularly relevant in the millimeter-wave (mmWave) range. In such scenarios, it is essential to account for the beam squint when attempting to recover channel parameters using a training sequence. However, the use of a training sequence alone may be insufficient for this purpose. To overcome this issue, in this work, we propose a channel estimation approach that exploits information provided by the control module of the vehicle, namely its velocity. The estimation problem that is designed, regards the channel both in a parametric and a non-parametric form and the alternating direction method of multipliers is utilised to efficiently solve it. It is demonstrated via simulations that considerable gains can be achieved if information from the control unit of the vehicle can be appropriately introduced and exploited. © 2023 European Signal Processing Conference, EUSIPCO. All rights reserved.

关键词： Unmanned aerial vehicles (UAV)

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：