检索结果-内蒙古大学图书馆

20th Joint ACL - ISO Workshop on Interoperable Semantic Annotation, ISA 2024

作者： Meeus, Quentin Moens, Marie-Francine Van Hamme, Hugo LIIR Lab. Computer Science Dpt. KU Leuven Belgium PSI Electrical Engineering Dpt. KU Leuven Belgium

ISBN: (纸本)9782493814326

While extensively explored in text-based tasks, Named Entity Recognition (NER) remains largely neglected in spoken language understanding. Existing resources are limited to a single, English-only dataset. This paper addresses this gap by introducing MSNER, a freely available, multilingual speech corpus annotated with named entities. It provides annotations to the VoxPopuli dataset in four languages (Dutch, French, German, and Spanish). We have also releasing an efficient annotation tool that leverages automatic pre-annotations for faster manual refinement. This results in 590 and 15 hours of silver-annotated speech for training and validation, alongside a 17-hour, manually-annotated evaluation set. We further provide an analysis comparing silver and gold annotations. Finally, we present baseline NER models to stimulate further research on this newly available dataset. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Silver

来源：评论

学校读者我要写书评

暂无评论

Whisper-Slu: Extending a Pretrained Speech-to-Text Transformer for Low Resource Spoken Language Understanding

Whisper-Slu: Extending a Pretrained Speech-to-Text Transform...

引用

2023 IEEE Automatic Speech Recognition and Understanding Workshop, ASRU 2023

作者： Meeus, Quentin Moens, Marie-Francine Van Hamme, Hugo Ku Leuven Dpt. Computer Science Belgium Ku Leuven Dpt. Electrical Engineering Belgium

ISBN: (纸本)9798350306897

Human-computer interactions require systems that work out of the box without requiring lots of data to adapt to a new task or user. In this research, we address low resource spoken language understanding tasks such as named entity recognition (NER), intent recognition (IR), and slot filling (SF) to research how a pretrained model can be modified for a new task, then finetuned with few labelled data. We propose extending the Whisper model with task-specific modules for NER, SF, and IR, leveraging a Markov network as output structure. We develop a novel approach to finetuning by removing irrelevant weights and reorganizing the embeddings to drastically improve the performance in a low-resource setting. Our approach outperforms previous models without external language models and demonstrates effective transfer learning, even with very limited training data. The models exhibit a small footprint, making them suitable for applications requiring robustness, few-shot learning, and efficiency. © 2023 IEEE.

关键词： Character recognition

来源：评论

学校读者我要写书评

暂无评论

Toward Secure Consumer IoT Communications using DNA Encryption and Blockchain Technology 6

Toward Secure Consumer IoT Communications using DNA Encrypti...

引用

6th International Conference on Pattern Analysis and Intelligent Systems, PAIS 2024

作者： Verma, Ankush Rathee, Geetanjali Kerrache, Chaker Abdelaziz Netaji Subhas University of Technology Dpt. of Computer Science And Engineering New Delhi110078 India Université Amar Telidji De Laghouat Dpt. of Computer Science Laghouat Algeria

ISBN: (纸本)9798350350265

Recently, the Internet of Things (IoT) has gained widespread popularity, yet its security remains a critical challenge due to the massive amount of information generated by connected devices. While encryption algorithms are commonly employed to protect this information, they are susceptible to various attacks. This paper introduces a novel encryption method to enhance data security in the Consumer IoT domain. The proposed approach combines DNA-inspired encryption with blockchain technology. The DNA encryption algorithm is utilized for the primary encryption phase, followed by the secure handling of encrypted data using blockchain. To evaluate the proposed scheme, a comparison is made with traditional techniques such as RSA, DNA, and a hybrid of both, considering encryption and decryption times. Simulations are conducted using MATLAB software, and an analytical study is performed to assess the performance of the suggested scheme. © 2024 IEEE.

关键词： Cryptography

来源：评论

学校读者我要写书评

暂无评论

A robust Carrier Frequency and Timing Offset Estimation in Inter-UAV OFDM Communication Links 7

A robust Carrier Frequency and Timing Offset Estimation in I...

引用

7th International Multi-Topic ICT Conference: IMTIC 2023

作者： Kulsoom, Farzana Fatima, Iraj Chaudhry, Hassan Nazeer Gul, Aneeqa Akram, Adeel University of Engineering and Technology Telecommunication Engineering Dpt. Taxila Pakistan Barani Institute of Technology Computer Science Dpt. Rawalpindi Pakistan

ISBN: (纸本)9798350338461

The UAV-mounted base station has been integrated into 5G and beyond to provide communication in an emergency when conventional systems fail due to some reason. Moreover, it is also used in coverage enhancement and for other strategic purposes. In this work, we have considered a MIMO-OFDM based UAV system that is inclined to fast-fading channels due to the continuous motion of UAVs. Due to the mobility of UAVs, there is a frequency and timing offset which must be estimated for proper and seamless communication. For that purpose, we have proposed a data-aided maximum likelihood-based scheme that efficiently estimates the carrier frequency and timing offset. Results show that our algorithm outperforms the state-of-the-art system in terms of MSE. © 2023 IEEE.

关键词： Orthogonal frequency division multiplexing

来源：评论

学校读者我要写书评

暂无评论

Towards an Efficient and Secure Cache Management Using Apriori-Based Interests Prediction in Named Data Networking 11

Towards an Efficient and Secure Cache Management Using Aprio...

引用

11th International Conference on Intelligent Systems and Embedded Design, ISED 2023

作者： Kerrache, Chaker Abdelaziz Rathee, Geetanjali Guellouma, Younes Gasmi, Hocine Ziani, Benameur Université Amar Telidji de Laghouat Dpt. of Computer Science Laghouat Algeria Netaji Subhas University of Technology Dpt. of Computer Science and Engineering New Delhi110078 India

ISBN: (纸本)9798350318159

Cache management is an important component in any network and it has even more importance in the Future Internet Architectures (FIAs) including Named Data Networking (NDN), because the caches play the key role in reducing the overall network latency and scalability. In this paper, we discuss the functionality of cache management in NDN, its types as well as its importance for the NDN architecture. In addition, we propose a machine learning-empowered cache management and interests predication for NDN to only preserve the cache only to the secure and really needed data. Our proposal uses Apriori algorithm which is supervised learning algorithm to find the association rules and then to recommend the next requested data. Implementation and experiments on real data traffic depicted that the network' performance and its influence on the cache increased by 3.2% for two content store sizes of 20 and 40 MB. In addition, a larger cache size of 80 MB shows an increase of the cache hit ratio reaching 90% and hence, clearly reducing the network latency. © 2023 IEEE.

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

Implementing a Blockchain Based System for Healthcare Applications Using Digital Twin 11

Implementing a Blockchain Based System for Healthcare Applic...

引用

11th International Conference on Intelligent Systems and Embedded Design, ISED 2023

作者： Benattia, Bouchra Zinab Houda Maicha, Nour El Kerrache, Chaker Abdelaziz Rathee, Geetanjali Calafate, Carlos T. Université Amar Telidji de Laghouat Dpt. of Computer Science Laghouat Algeria Netaji Subhas University of Technology Dpt. of Computer Science and Engineering New Delhi110078 India Universitat Politècnica de València Dpt. of Computer Engineering València Spain

ISBN: (纸本)9798350318159

With the increasing focus on healthcare, especially real-time analytics and self-diagnosis, the interest in capturing real-time patient data has increased significantly for both physicians and patients. The development of smart devices that can collect data such as blood pressure, heart rate, blood sugar level, etc., has greatly contributed to improving individual health management capabilities. However, these advances also, bring challenges such as concerns over safety, reliability, and accurate diagnosis. Problems often arise, especially when exchanging information and transmitting data. To overcome these challenges, this project uses blockchain technology to improve data transmission and security. This study explores the use of blockchain-based internet solutions by integrating blockchain into smart health-care. Additionally, the digital twin concept is used to improve diagnosis by simulating the effects of drugs on virtual objects. Based on the results of these simulations, physicians can make informed decisions about appropriate medications and accurately assess a patient's condition. A health monitoring system using digital twin technology and blockchain can be implemented to securely organize the collection of user information and enable accurate diagnosis. This integration ensures secure storage and transmission of your data and improves your health outcomes. In this paper, we study the integrating blockchain technology with Digital Twin for patient health monitoring. We showcase an interface connected to the blockchain database, specifically designed for doctors to review patient health records through patient wallet address and determine the normalcy of the patient's heart rate based on their medical data. Additionally, the interface allows doctors to predict the effect of medication on the patient's heart rate and blood pressure. © 2023 IEEE.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Enhancing Concept Drift Detection in Drifting and Imbalanced Data Streams through Meta-Learning

Enhancing Concept Drift Detection in Drifting and Imbalanced...

引用

2023 IEEE International Conference on Big Data, BigData 2023

作者： Aguiar, Gabriel Jonas Cano, Alberto Virginia Commonwealth University Dpt. of Computer Science College of Engineering RichmondVA United States

ISBN: (纸本)9798350324457

One of the biggest challenges in learning from data streams is adapting the classification model to new data. Due to the evolving nature of data streams, they are subject to a phenomenon known as concept drift that makes previously learned knowledge and model outdated. Therefore, concept drift must be efficiently detected in order to adapt the classification model. While there exists a plethora of drift detectors, with different mechanisms, selecting the most suitable for a new stream is a difficult task, since apriori knowledge may not be available and changes over time can affect the performance of the detector. This paper proposes a framework that exploits statistical and temporal meta-features from sliding windows to dynamically recommend a suitable drift detector in real-time for unseen chunks of streams according to its properties using Meta-Learning. We performed experiments on 10 real-world data streams and 18 synthetic generated data streams that were subject to concept drift and class imbalance in order to evaluate the performance of the proposed framework. Experiments exposed that the proposed approach was able to enhance the concept drift detection in a variety of scenarios demonstrating robustness to class imbalance and the advantages of dynamically selecting the drift detector. © 2023 IEEE.

关键词： Concept Drift Data streams Machine Learning Meta-Learning

来源：评论

学校读者我要写书评

暂无评论

Leveraging protein-protein interactions in phenotype prediction through graph neural networks 21

Leveraging protein-protein interactions in phenotype predict...

引用

21st IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, CIBCB 2024

作者： Smeriglio, Riccardo Rosell-Mirmi, Joana Radeva, Petia Abante, Jordi Universitat de Barcelona Dpt. of Physics Spain Politecnico di Torino Control and Computer Engineering Dpt. Turin Italy Headache Research Group Vall d'Hebron Institut de Recerca Catalonia Barcelona Spain Universitat de Barcelona Institute of Neurosciences Dpt. of Mathematics & Computer Science Catalonia Barcelona Spain Laboratory of Stem Cells & Regenerative Medicine Creatio Production & Validation Center of Advanced Therapies Spain Universitat de Barcelona Institute of Neurosciences Dpt. of Biomedical Sciences Spain Dpt. of Mathematics & Computer Science Catalonia Barcelona Spain

ISBN: (纸本)9798350356632

Current genotype-to-phenotype models, such as polygenic risk scores, only account for linear relationships between genotype and phenotype and ignore epistatic interactions, limiting the complexity of the diseases that can be properly characterized. Protein-protein interaction networks have the potential to improve the performance of the models. Moreover, interactions at the protein level can have profound implications in understanding the genetic etiology of diseases and, in turn, for drug development. In this article, we propose a novel approach for phenotype prediction based on graph neural networks (GNNs) that naturally incorporates existing protein interaction networks into the model. As a result, our approach can naturally discover relevant epistatic interactions. We assess the potential of this approach using simulations and comparing it to linear and other non-linear approaches. We also study the performance of the proposed GNN-based methods in predicting Alzheimer's disease, one of the most complex neurodegenerative diseases, where our GNN approach outperform state of the art methods. In addition, we show that our proposal is able to discover critical interactions in the Alzheimer's disease. Our findings highlight the potential of GNNs in predicting phenotypes and discovering the underlying mechanisms of complex diseases. © 2024 IEEE.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

FraMark: A Blockchain Marketplace for a 5G Network Management using Fractional NFTs 7

FraMark: A Blockchain Marketplace for a 5G Network Managemen...

引用

7th World Symposium on Communication engineering, WSCE 2024

作者： Michoulis, George Tsourdinis, Kyros Niotis, George Kyranou, Konstantinos Fountoukidis, Eleutherios Lagkas, Thomas Argyriou, Vasileios Sarigiannidis, Panagiotis Psannis, Konstantinos E. Goudos, Sotirios K. Sidroco Holdings Ltd R&D Dpt. Thessaloniki Greece Sidroco Holdings Ltd R&D Dpt. Nicosia Cyprus Democritus University of Thrace Dpt. of Computer Science Kavala Greece University of Western Macedonia Dpt. of Electrical & Computer Engineering Kozani Greece Kingston University London Kingston Upon Thames Dpt. of Networks and Digital Media Surrey United Kingdom University of Macedonia Department of Applied Informatics Thessaloniki Greece ELEDIA@AUTH School of Physics Aristotle University of Thessaloniki Thessaloniki Greece

ISBN: (纸本)9798331542825

This paper presents FraMark, a blockchain-based framework for 5G network management that utilizes Hyperledger Fabric and fractional Non-Fungible Tokens (NFTs) to optimize resource allocation. We introduce a comprehensive system that fractionalizes digital assets to represent 5G network resources, enhancing efficiency and allocation flexibility. Our framework addresses the challenges of 5G network management by ensuring low latency, high transaction throughput, and secure resource trading. We conduct experiments to evaluate the performance of our approach, including CPU and memory utilization, latency analysis, and block size assessment. The results demonstrate the effectiveness of FraMark in improving network resource utilization and transaction efficiency, validating our framework's claims. By leveraging blockchain technology and innovative resource slicing techniques, FraMark paves the way for dynamic and efficient 5G network management. © 2024 IEEE.

关键词： 5G mobile communication systems

来源：评论

学校读者我要写书评

暂无评论

MSNER: A Multilingual Speech Dataset for Named Entity Recognition

arXiv

引用

arXiv 2024年

作者： Meeus, Quentin Moens, Marie-Francine Van Hamme, Hugo LIIR Lab Computer Science Dpt. KU Leuven Belgium PSI Electrical Engineering Dpt. KU Leuven Belgium

关键词： Silver

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：