检索结果-内蒙古大学图书馆

International Conference on Circuit, Power and Computing Technologies (ICCPCT)

作者： Sanjana Prasad Ishu Sharma Deepashree Rajendraprasad Department of Electronics and Communication Engineering HKBK College of Engineering Bangalore India Department of Computer Science and Engineering Lovely Professional University Phagwara Punjab India Department of Health Informatics Boston University Boston Massachusetts USA

ISBN: (数字)9798350372816

ISBN: (纸本)9798350372823

The Industrial Internet of Things has emerged as an essential tool for building Industry 4.0 and Industry 5.0 where timely information can be retrieved from different scenarios. These devices are highly vulnerable to cyberattacks as heterogeneous types of devices can be used in the infrastructure that may or may not be equipped with standardized security protocols. Artificial intelligence-based methodologies can effectively identify these types of attacks on a prior basis for taking mitigation action. This method raises concerns about data privacy as building a machine learning-based method requires the sharing of network data that can reveal the actual information of industries. The proposed Federated Learning based framework handles this concern by preserving each device’s critical data by utilizing the benefits of the decentralized model aggregation. This research work presents the comparison of the proposed framework on the CICIoT2023 dataset with federated averaging and federated proximal techniques for achieving a global model. The performance evaluation of these two aggregation techniques is performed based on metrics of accuracy, loss, precision and recall. The results prove that the federated proximal method achieves higher accuracy in comparison to the federated averaging method.

关键词： Performance evaluation Learning systems Accuracy Protocols Federated learning Prevention and mitigation Buildings

来源：评论

学校读者我要写书评

暂无评论

Throughput Analysis of SIC-Based Two-Device Slotted ALOHA with Feedback Over Nakagami-m Fading Channels 18

Throughput Analysis of SIC-Based Two-Device Slotted ALOHA wi...

引用

18th International Symposium on Information Theory and Its Applications, ISITA 2024

作者： Fukui, Daiki Takahashi, Yuhei Song, Guanghui Kimura, Tomotaka Liu, Zilong Cheng, Jun Doshisha University Department of Intelligent Information Engineering and Sciences Kyoto610-0321 Japan Xidian University The State Key Laboratory of Integrated Services Networks Xi'an710071 China School of Computer Science and Electronic Engineering University of Essex ColchesterCO4 3SQ United Kingdom

ISBN: (纸本)9784885523519

Throughput analysis for successive interference cancellation-based two-device slotted ALOHA with feedback is studied over Nakagami-m fading channels. Explicit expressions for the state transition probabilities are derived for a Markov process, thus facilitating the computation of the throughput. Through optimization of the transmission probability, it is shown that the maximum throughput is achieved for a given code rate. © 2024 The Institute of electronics, Information and Communication Engineers.

关键词： Markov processes

来源：评论

学校读者我要写书评

暂无评论

Trend and Correlation Analysis of Instagram Activity Using Data Mining and Statistics 6

Trend and Correlation Analysis of Instagram Activity Using D...

引用

6th International Conference on Cybernetics and Intelligent System, ICORIS 2024

作者： Musu, Wilem Samsie, Indra Fawait, Aldi Bastiatul Lempan, Nadia Nurliah Gabriel, Maechel Maximiliano Sinar Department of Informatics Engineering Dipa Makassar University Makassar Indonesia Department of Information System Dipa Makassar University Makassar Indonesia Department of Computer Science Widya Gama Mahakan University Samarinda Indonesia Department of Digital Business Dipa Makassar University Makassar Indonesia Department of Entrepreneurship Dipa Makassar University Makassar Indonesia

ISBN: (纸本)9798331539603

With the growing importance of social media for brand engagement, there remains a limited understanding of how specific metrics—such as Reach, Followers, and New Followers—impact Instagram user interaction over time. This research seeks to address this gap by examining which variables most effectively drive Engagement on Instagram, providing insights for strategic content optimization. Through an analysis of 100 posts, using multiple linear regression and decision tree methods, the study investigates the relationship between Reach, Followers, New Followers, and Engagement. Results indicate that Reach, Followers, and New Followers are not statistically significant predictors of Engagement, as evidenced by high p-values (all > 0.05). Decision tree results demonstrate a Mean Absolute Error (MAE) of 0.0226 and an R2 Score of 0.8003, suggesting moderate predictive accuracy but with potential for refinement. Moving average analysis reveals fluctuating Engagement with no stable long-term trend. A correlation matrix further shows that Reach and New Followers exhibit high correlations with Engagement (0.9723 and 0.9707, respectively), while Followers show a weak correlation (0.2575). These findings highlight the limited impact of Followers on Engagement, underscoring the importance of Reach and New Followers. This study contributes novel insights by identifying key engagement drivers, emphasizing the need for more targeted strategies in Instagram campaign planning to enhance user interaction. © 2024 IEEE.

关键词： Multiple linear regression

来源：评论

学校读者我要写书评

暂无评论

Dynamic Shortening of Psychometric Questionnaires: A Focus on PHQ-9 12th

Dynamic Shortening of Psychometric Questionnaires: A Focus o...

引用

12th Scientific Meeting on International Society for Research on Internet Interventions, ISRII-12 2024

作者： Abdulhussein, Zahraa Van De Ven, Pepijn Department of Electronic and Computer Engineering University of Limerick LimerickV94 T9PX Ireland Health Research Institute University of Limerick V94 T9PX Ireland

The Patient Health Questionnaire-9 (PHQ-9) is a widely utilized assessment tool designed to screen for and measure the severity of depression. Recent research has explored the viability of shorter versions of the questionnaire, such as the PHQ-DEP-4 and PHQ-2, particularly for screening depressed individuals before clinical trials. However, these short tests use a fixed number of questions. This study investigated whether there is merit in posing a dynamic number of questions in screening individuals for depressive symptomatology. To address this, our study focuses on creating a short and dynamic version of the PHQ-9. Our approach utilizes dynamic thresholds tailored to individual responses to predict depression. Optimal thresholds are determined by minimizing the number of questions while simultaneously maximizing the Youden index. Results indicate the Dynamic PHQ-9 model demonstrated superior discrimination ability achieving highest Youden index, sensitivity, and specificity. © 2024 The Authors.

关键词： Clinical research

来源：评论

学校读者我要写书评

暂无评论

Online Learning to Cache and Recommend in the Next Generation Cellular Networks

IEEE Transactions on Machine Learning in Communications and ...

引用

IEEE Transactions on Machine Learning in Communications and Networking 2024年 2卷 511-525页

作者： Tharakan, Krishnendu S Bharath, B.N. Bhatia, Vimal Queen's University School of Computing KingstonONK7L 3N6 Canada Dharwad580011 India Indian Institute of Technology Indore Department of Electrical Engineering Center for Advanced Electronics Indore453552 India University of Hradec Králové Faculty of Informatics and Management Hradec Králové500 03 Czech Republic Soochow University School of Electronic and Information Engineering Suzhou215006 China

An efficient caching can be achieved by predicting the popularity of the files accurately. It is well known that the popularity of a file can be nudged by using recommendation, and hence it can be estimated accurately leading to an efficient caching strategy. Motivated by this, in this paper, we consider the problem of joint caching and recommendation in a 5G and beyond heterogeneous network. We model the influence of recommendation on demands by a Probability Transition Matrix (PTM). The proposed framework consists of estimating the PTM and use them to jointly recommend and cache the files. In particular, this paper considers two estimation methods namely a) Bayesian estimation and b) a genie aided Point estimation. An approximate high probability bound on the regret of both the estimation methods are provided. Using this result, we show that the approximate regret achieved by the genie aided Point estimation approach is mathcal {O}(T{2/3} sqrt {log T}) while the Bayesian estimation method achieves a much better scaling of mathcal {O}(sqrt {T}). These results are extended to a heterogeneous network consisting of M small base stations (SBSs) with a central macro base station. The estimates are available at multiple SBSs, and are combined using appropriate weights. Insights on the choice of these weights are provided by using the derived approximate regret bound in the multiple SBS case. Finally, simulation results confirm the superiority of the proposed algorithms in terms of average cache hit rate, delay and throughput. © 2023 CCBY.

关键词： Heterogeneous networks

来源：评论

学校读者我要写书评

暂无评论

Detection of Katokkon Chili Maturity using Convolutional Neural Network with Transfer Learning Model DenseNet169

Detection of Katokkon Chili Maturity using Convolutional Neu...

引用

Artificial Intelligence and Image Processing (IWAIIP), International Workshop on

作者： Andi Baso Kaswar Muhammad Ryan Ilham Renisa Amalia Tahir Abdul Muis Mappalotteng Yasser Abdul Djawad Dyah Darma Andayani Department of Computer Engineering State University of Makassar Makassar Indonesia Department of Informatics and Computer Engineering State University of Makassar Makassar Indonesia Department of Electronics Engineering State University of Makassar Makassar Indonesia

Chili is a plant that originates from America and is included in the Solonaceae and Capsicum genera. One type of chili that is included in the large chili category is Katokkon chili, commonly referred to as Toraja chili, which comes from the Tana Toraja district, South Sulawesi, Indonesia. Research related to the detection of fruit ripeness and quality has also been conducted. However, the research that has been conducted still uses relatively few images, and the accuracy needs to be improved. In addition, no research has discussed the detection of Katokkon chili maturity level using images or images of Katokkon chili fruit taken directly from the tree. Therefore, research was conducted with the title Detection of the Maturity Level of Katokkon Chili using the CNN pretrained DenseNet169 model method. In this study, the level of maturity of chili peppers was classified into three categories: raw chilies, half-ripe chilies, and ripe chilies. Based on the results of several tests that have been carried out, the best accuracy results are obtained in testing using the DenseNet169 method, with an accuracy of 95.03%, a loss value of 16.72%, a recall and specificity value of 95.03% and 97.51% respectively, a precision value of 95.03%, and an accuracy CF of 96.69%. This shows that the DenseNet169 method with a combination (Swish, Swish, and Adamax) can be considered the best choice for classifying or detecting the level of maturity of Toraja chili compared other methods that have been tested.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Domain Generalization with fourier Transform and soft thresholding

Domain Generalization with fourier Transform and soft thresh...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Hongyi Pan Bin Wang Zheyuan Zhang Xin Zhu Debesh Jha Ahmet Enis Cetin Concetto Spampinato Ulas Bagci Department of Radiology Machine & Hybrid Intelligence Lab Northwestern University USA Department of Electrical and Computer Engineering University of Illinois Chicago USA Department of Electrical Electronic and Computer Engineering University of Catania Italy

Domain generalization aims to train models on multiple source domains so that they can generalize well to unseen target domains. Among many domain generalization methods, Fourier-transformbased domain generalization methods have gained popularity primarily because they exploit the power of Fourier transformation to capture essential patterns and regularities in the data, making the model more robust to domain shifts. The mainstream Fouriertransform-based domain generalization swaps the Fourier amplitude spectrum while preserving the phase spectrum between the source and the target images. However, it neglects background interference in the amplitude spectrum. To overcome this limitation, we introduce a soft-thresholding function in the Fourier domain. We apply this newly designed algorithm to retinal fundus image segmentation, which is important for diagnosing ocular diseases but the neural network’s performance can degrade across different sources due to domain shifts. The proposed technique basically enhances fundus image augmentation by eliminating small values in the Fourier domain and providing better generalization. The innovative nature of the soft thresholding fused with Fourier-transform-based domain generalization improves neural network models’ performance by reducing the target images’ background interference significantly. Experiments on public data validate our approach’s effectiveness over conventional and state-of-the-art methods with superior segmentation metrics.

关键词：

来源：评论

学校读者我要写书评

暂无评论

MetaCon: Revitalizing Internet Congestion Control with Meta-Reinforcement Learning

MetaCon: Revitalizing Internet Congestion Control with Meta-...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： He Bai Hui Li Jianming Que Minglong Zhang Peter Han Joo Chong Kalupahana Liyanage Kushan Sudheera Xinyuan Pei School of Electronic and Computer Engineering Peking University Shenzhen China School of Computer Science Fuyao University of Science and Technology Fuzhou China Department of Electrical and Computer Engineering Mississippi State University Starkville USA School of Engineering Computer and Mathematical Sciences Auckland University of Technology Auckland New Zealand Department of Electrical and Information Engineering Faculty of Engineering University of Ruhuna Sri Lanka

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Effective congestion control algorithms (CCAs) are crucial for the smooth operation of Internet communication infrastructure. CCAs adjust transmission rates based on congestion signals, optimizing resource utilization and user experience. However, existing studies, both rule-based and learning-based CCAs, often struggle with generalization and underperform when deployed in real-world environments. When applied to unseen network conditions, hand-crafted schemes or pre-trained models may experience significant performance degradation. To address this challenge, we propose MetaCon, a novel adaptive Internet congestion control approach based on meta-reinforcement learning. MetaCon leverages knowledge learned from prior scenarios to quickly adapt to new environments. Experimental results show that MetaCon outperforms existing algorithms by exhibiting superior generalization and achieving better transmission performance across a wide variety of network conditions.

关键词： Degradation Adaptation models Signal processing algorithms Signal processing User experience Acoustics Internet Resource management Speech processing

来源：评论

学校读者我要写书评

暂无评论

Design and Analysis of Terahertz Multi-Frequency Low-Profile Waveguide Antenna Based on Micro-Coaxial Feeding

Design and Analysis of Terahertz Multi-Frequency Low-Profile...

引用

Antennas and Propagation & USNC/URSI National Radio Science Meeting

作者： LiBo Wang Shu Lin Tong Xu Xingqi Zhang Xinyue Zhang School of Electronics and Information Engineering Harbin Institute of Technology Harbin China Department of Electrical and Computer Engineering University of Alberta Edmonton Canada Department of Electronic Engineering Maynooth University Ireland

ISBN: (数字)9798350369908

ISBN: (纸本)9798350369915

This paper presents a comprehensive theoretical analysis of electric field distribution in low-profile waveguide antennas operating in the Terahertz frequency range. We have developed a micro-coaxial-based, low-profile, multi-frequency waveguide antenna, integrating pin and gradient ridge structures to enhance its performance. The antenna dimensions are 4.7 mm ×13.3 mm × 0.9 mm. The reflection coefficient of this waveguide antenna is below -10 dB within the frequency ranges of 110–130 GHz, 148–156 GHz, and 175–182 GHz. It exhibits excellent directional radiation characteristics at the center frequencies of these three bands. With its compact size, lightweight, and low-profile features, this antenna is suitable for Terahertz high-speed communication applications, providing a practical approach for harnessing the electromagnetic spectrum resources in high-frequency domains.

关键词： Antenna theory Conferences Waveguide theory Reflector antennas Antenna feeds Reflection Reflection coefficient

来源：评论

学校读者我要写书评

暂无评论

Improving DL-Learner on a Malware Detection Use Case 36

Improving DL-Learner on a Malware Detection Use Case

引用

36th International Workshop on Description Logics, DL 2023

作者： Bisták, Tomáš Švec, Peter Kluka, Ján Šimko, Alexander Balogh, Štefan Homola, Martin Department of Applied Informatics Faculty of Mathematics Physics and Informatics Comenius University Mlynská dolina Bratislava842 48 Slovakia Institute of Computer Science and Mathematics Faculty of Electrical Engineering and Information Technology Slovak University of Technology Ilkovicova 3 Bratislava812 19 Slovakia

Automation of malware characterization has become increasingly important for early malware detection over the past decades. Since it is crucial to be able to perform malware detection transparently, explainable machine-learning methods may be considered particularly suitable for this application. In our previous work, we thus employed algorithms for learning description-logics concept expressions to obtain structured characterizations of malicious software. The results of our initial experiments, however, lead us to several inconsistencies in the behaviour of the studied algorithms included in DL-Learner, a widely adopted framework for concept learning. Hence, in this work, we make a couple of corrections to DL-Learner, along with a few further enhancements, and repeat the experiments to empirically assess the effects of our changes. The outcomes show that the modified algorithms can mark an improvement in terms of the predictability of their behaviour as well as evaluation metrics. © 2023 CEUR-WS. All rights reserved.

关键词： Malware

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：