检索结果-内蒙古大学图书馆

International Conference on Radioelektronika

作者： Jozef Badar Jan Papaj Department of Electronics and Multimedia Telecommunications Faculty of Electrical Engineering and Informatics Technical University of Košice Košice Slovakia

ISBN: (数字)9798331544478

ISBN: (纸本)9798331544485

Mobile ad hoc networks (MANETs) form the foundation of dynamic and decentralized wireless communication. Although existing simulators, such as NS-2, NS-3, and other widely used tools, offer various modeling capabilities, many of them encounter limitations in implementing advanced AI-based routing protocols, including high configuration complexity, lower scalability, and limited real-time visualization capabilities. This paper presents a custom MANET simulator supporting both traditional and AI-based routing protocols, designed with an emphasis on scalability, modularity, and real-time network visualization, while also considering energy modeling and dynamic obstacles. Experimental results indicate that AI-based algorithms, such as Ant Colony Optimization (ACO), can achieve higher packet delivery ratio (PDR), better route stability, and improved energy efficiency compared to traditional protocols. These findings suggest significant potential for utilizing AI techniques in MANET routing, particularly in dynamically changing environments.

关键词： Visualization Scalability Routing Routing protocols Ad hoc networks Real-time systems Stability analysis Energy efficiency Mobile computing AODV

来源：评论

学校读者我要写书评

暂无评论

Grammatical Error Correction in Slovak Using Neural Machine Translation and No Language Left Behind

Grammatical Error Correction in Slovak Using Neural Machine ...

引用

International Conference on Radioelektronika

作者： Maroš Harahus Zuzana Sokolová Matúš Pleva Miroslav Sokol Department of Electronics and Multimedia Communications Faculty of Electrical Engineering and Informatics Technical University of Košice Košice Slovakia

ISBN: (数字)9798331544478

ISBN: (纸本)9798331544485

Grammatical error correction aims to identify and rectify linguistic inaccuracies in text, thereby enhancing communication and comprehension. This study investigates the ap-plication of neural machine translation techniques, particularly the No Language Left Behind technique for GEC in Slovak, a morphologically rich language. We constructed datasets containing deliberate grammatical errors and used them to train models capable of detecting and correcting these errors. By introducing random character and word-level transformations, we simulated realistic grammatical errors, creating a robust testing ground for model evaluation. Our experimental results demonstrate significant improvements in performance metrics such as BLEU score, Word Error Rate, and Sentence Error Rate. The findings highlight the potential of NMT-based approaches in developing efficient and accurate GEC systems for underrepresented languages.

关键词： Accuracy Error analysis Statistical analysis Neural machine translation Machine learning Linguistics Robustness Error correction

来源：评论

学校读者我要写书评

暂无评论

Sentiment Analysis Using Transformer Models: BERT, T5, and GPT

Sentiment Analysis Using Transformer Models: BERT, T5, and G...

引用

International Conference on Radioelektronika

作者： Zuzana Sokolová Maroš Harahus Miroslav Sokol Eva Kupcová Matúš Pleva Department of Electronics and Multimedia Communications Faculty of Electrical Engineering and Informatics Technical University of Košice Košice Slovakia

ISBN: (数字)9798331544478

ISBN: (纸本)9798331544485

Sentiment analysis, or opinion mining, is a crucial Natural Language Processing task that classifies text into sentiment categories such as positive, negative, or neutral. While early approaches relied on lexicon-based methods and traditional machine learning models, recent advances in deep learning, particularly transformer-based architectures, have significantly improved sentiment classification performance. This paper investigates the application of state-of-the-art transformer models to sentiment analysis, with a focus on Slovak-language datasets. We evaluate their effectiveness using three datasets: SentiSK, Sentigrade, and the Slovak dataset for SA. Our experiments include both three-class and two-class classification tasks, comparing the performance of RoBERTa, DistilBERT, mT5, byT5, and GPT models based on accuracy, F1-score, precision, and recall. The results demonstrate that GPT and mT5 achieve the highest performance across various datasets, while DistilBERT shows strong potential for specific tasks. Our findings highlight the advantages of transformer-based models for sentiment analysis and emphasize the importance of dataset-specific optimizations.

关键词： Measurement Deep learning Sentiment analysis Analytical models Accuracy Text categorization Transformers Data augmentation Data models Optimization

来源：评论

学校读者我要写书评

暂无评论

Comparison of Feature Extraction Algorithms for Speech-Based Emotional Stress Level Detection

Comparison of Feature Extraction Algorithms for Speech-Based...

引用

International Conference on Radioelektronika

作者： Ján Staš Katarina Škotková Lukáš Hudák Department of Electronics and Multimedia Communications Faculty of Electrical Engineering and Informatics Technical University of Košice Košice Slovakia

ISBN: (数字)9798331544478

ISBN: (纸本)9798331544485

With the increasing use of human health monitoring applications, the detection of emotional stress from speech has become an important research topic. This study highlights the effectiveness of various feature extraction algorithms in classifying speech segments into three stress levels, emphasizing the superior performance of gammatone cepstral coefficients (GTCC), particularly when combined with discrete wavelet transform (DWT). The experimental results of the Fl-score of 80.8% on average confirm that an advanced combination of features and ensemble classifiers, such as Bagged Trees or Subspace k-Nearest Neighbor, can significantly enhance classification accuracy.

关键词： Deep learning Anxiety disorders Transforms Nearest neighbor methods Feature extraction Transformers Discrete wavelet transforms Classification algorithms Speech processing Monitoring

来源：评论

学校读者我要写书评

暂无评论

Automatic Speech Liveness Detection

Automatic Speech Liveness Detection

引用

International Conference on Emerging eLearning Technologies and Applications (ICETA)

作者： Renát Haluška Eva Kupcová Matáš Pleva František Hric Department of Electronics and Multimedia Communications Faculty of Electrical Engineering and Informatics Technical University of Kosice Koáice Slovakia

ISBN: (数字)9798331527716

ISBN: (纸本)9798331527723

This paper addresses the growing challenge of detecting deepfake audio, which can mimic or alter speech using advanced machine learning techniques like GAN sand RNNs. Such audio manipulation poses risks in biometric security and fraud. The study focuses on developing detection methods using acoustic features such as Mel-Frequency Cepstral Coefficients (MFCCs) and zero-crossing rate (ZCR). Two neural network models were tested, with the best achieving accuracy in identifying synthetic audio. The results highlight the need for advanced detection algorithms to keep pace with evolving deepfake technologies, enhancing biometric and digital security systems.

关键词： Biometrics Deepfakes Visualization Accuracy Cepstral analysis Biological system modeling Machine learning Feature extraction Fraud Security

来源：评论

学校读者我要写书评

暂无评论

Optimizing Optical Attocells Positioning of Indoor Visible Light Communication System

引用

Computers, Materials & Continua 2023年第2期74卷 3607-3625页

作者： Mohammed S.M.Gismalla Asrul I.Azmi Mohd R.Salim Farabi Iqbal Mohammad F.L.Abdullah Mosab Hamdan Muzaffar Hamzah Abu Sahmah M.Supa’at School of Electrical Engineering Faculty of EngineeringUniversiti Teknologi MalaysiaSkudai81310Malaysia Lightwave Communications Research Group(LCRG) Faculty of Electrical EngineeringUniversiti Teknologi Malaysia(UTM)SkudaiJohorMalaysia Department of Electronics and Electrical Engineering Faculty of EngineeringInternational University of AfricaKhartoum12223Sudan Department of Communication Engineering Faculty of Electrical and Electronic Engineering Universiti Tun Hussein Onn Malaysia(UTHM)Parit RajaBatu Pahat86400JohorMalaysia Faculty of Computing and Informatics Universiti Malaysia SabahKota Kinabalu88400Malaysia Department of Computer Science University of São PauloSão PauloBrazil

Visible light communication(VLC),which is a prominent emerging solution that complements the radio frequency(RF)technology,exhibits the potential to meet the demands of fifth-generation(5G)and beyond *** random movement of mobile terminals in the indoor environment is a challenge in the VLC *** model of optical attocells has a critical role in the uniform distribution and the quality of communication links in terms of received power and signal-to-noise ratio(SNR).As such,the optical attocells positions were optimized in this study with a developed try and error(TE)*** optimized optical attocells were examined and compared with previous *** novel approach had successfully increased minimum received power from−1.29 to−0.225 dBm,along with enhanced SNR performance by 2.06 *** bit error rate(BER)was reduced to 4.42×10−8 and 6.63×10−14 by utilizing OOK-NRZ and BPSK modulation techniques,*** optimized attocells positions displayed better uniform distribution,as both received power and SNR performances improved by 0.45 and 0.026,*** the results of the proposed model are optimal,it is suitable for standard office and room model applications.

关键词： Visible light communication(VLC) optical attocell received power signal-to-noise ratio(SNR) bit error rate(BER) coefficient of variation(CV)

来源：评论

学校读者我要写书评

暂无评论

High-temperature Corrosion of ∼ 30 Pct Porous FeCr Stainless Steels in Air: Long-Term Evaluation Up to Breakaway (vol 54, pg 2244, 2023)

引用

METALLURGICAL AND MATERIALS TRANSACTIONS A-PHYSICAL METALLURGY AND MATERIALS SCIENCE 2023年第8期54卷 3370-3370页

作者： Koszelow, D. Advanced Materials Centre Faculty of Electronics Telecommunications and Informatics Gdańsk University of Technology 80-233 Gdańsk Poland Laboratory for Nuclear Materials Laboratory for Synchrotron Radiation and Femtochemistry Paul Scherrer Institut 5232 Villigen Switzerland International Centre of Electron Microscopy for Materials Science Faculty of Metal Engineering and Industrial Computer Science AGH University of Science and Technology Al. Mickiewicza 30 30-059 Krakow Poland

In this work, a long-term (up to 6000 hours) corrosion evaluation of three porous (~ 30 pct of initial porosity) ferritic iron-chromium alloys with different Cr contents (20, 22, and 27 wt pct of Cr) was carried out at 600 °C, 700 °C, 800 °C, and 900 °C in air. Mass gain measurements and SEM analyses revealed that at temperatures above 600 °C, all alloys exhibit breakaway corrosion, whereas at 600 °C, none of the alloys were heavily oxidized even after 6000 hours. Based on the results, the diffusion character of the corrosion of porous chromia-forming alloys was identified. The microstructure changes at high temperatures in porous alloys containing 22 wt pct of Cr were determined in detail by transmission electron microscopy. The proposed prediction model indicated that the lifetimes of the Fe20Cr and Fe22Cr alloys were determined as 1250 hours (± 535 hours) and 1460 hours (± 640 hours), respectively. It is in agreement with the long-term oxidation experiment. For the Fe27Cr alloy, the deviation between predicted and observed lifetimes occurs. The proposed model allows for qualitative estimation of the porous alloys’ lifetime with experimentally validated accuracy.

关键词： Chromium alloys

来源：评论

学校读者我要写书评

暂无评论

Facial Liveness Detection for Biometric User Identification

Facial Liveness Detection for Biometric User Identification

引用

International Conference on Radioelektronika

作者： Renát Haluška Matúš Pleva Jozef Sedlák Marianna Koctúrová Samuel Andrečík Ľuboš Ovseník Department of Electronics and Multimedia Communications Faculty of Electrical Engineering and Informatics Technical University Kosice Košice Slovakia

This scientific paper addresses the pervasive issue of spoofing in biometric user identification, focusing on its manifestation in facial recognition systems. Spoofing involves deceptive communication originating from an untrusted source, aiming to gain unauthorized access to sensitive information. The study delves into the vulnerabilities of facial biometrics, particularly in scenarios where malicious actors attempt to imitate a user’s face using masks, photos, or digital means.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Children-robot spoken interaction in selected educational scenarios 20

Children-robot spoken interaction in selected educational sc...

引用

20th IEEE International Conference on Emerging eLearning Technologies and Applications, ICETA 2022

作者： Ondas, S. Pleva, M. Juhar, J. Technical University of Kosice Faculty of Electrical Engineering and Informatics Dept. of Electronics and Multimedia Communications Letna 9 Košice04120 Slovakia

ISBN: (数字)9798350320336

ISBN: (纸本)9798350320336

The proposed paper brings experiences and analyzes observations collected during children-robot spoken interaction recording during Children's University 2022 at the Technical University of Košice. We conducted the experiment in Wizard-of-Oz scenario, which was focused on backchanneling in dialogue and on some interaction scenarios, which are similar to interaction scenarios between a teacher and children during education. © 2022 IEEE.

关键词： Educational robots

来源：评论

学校读者我要写书评

暂无评论

Improved-Reliability Antenna Optimization through Penalized Design Specification Scanning

引用

IEEE Antennas and Wireless Propagation Letters 2025年

作者： Koziel, Slawomir Pietrenko-Dabrowska, Anna Reykjavik University Engineering Optimization and Modeling Center Reykjavik Iceland Gdansk University of Technology Faculty of Electronics Telecommunications and Informatics Gdansk80-233 Poland

Antenna engineering has become reliant on rigorous optimization methods. Although local parameter tuning is generally preferred due to manageable costs, it is often insufficient in cases where the starting point is of poor quality. The utilization of global optimization techniques is impeded by extreme computational overhead. This letter proposes a simple yet efficient approach to enable globalized search while employing local techniques. Our methodology reformulates the optimization task by evaluating antenna performance regarding the target operating bandwidths scanned within a pre-defined frequency range and applying penalties proportional to the misalignment between the original and scanned targets. This approach makes the design specifications attainable through local parameter adjustment for considerably broader frequency spectra than conventional formulation, thereby enabling globalized optimization at little extra computational expenses. Our technique is demonstrated using several antennas and showcased to operate dependably and outperform standard gradient-based algorithms dramatically. © 2002-2011 IEEE.

关键词： Scanning antennas

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：