检索结果-内蒙古大学图书馆

e-Prime - Advances in Electrical engineering, Electronics and Energy 2025年 12卷

作者： Nazir, Amril Shaikh, Abdul Khalique Khan, Aftab Ahmed Shah, Abdul Salam Khalique, Nadia Department of Information Systems and Technology Management Zayed University Abu Dhabi 144534 United Arab Emirates Department of Information Systems Sultan Qaboos University Muscat Oman Department of Computer Science Karakoram International University Gilgit Pakistan School of Computer Science Faculty of Innovation and Technology Taylor's University Selangor Subang Jaya 47500 Malaysia

The increasing adoption of electric vehicles has led to the installation of charging stations in various locations in major cities worldwide. This study focuses on energy consumption forecasting for Boulder, Nevada, United States electric vehicle charging stations. Efficient management of energy resources at these charging points is crucial for optimizing resource utilization and reducing charging time. While existing literature has focused on energy consumption prediction in smart homes and grids, the significance of electric charging points in smart cities must be considered. The transformers have handled time series forecasting better with larger datasets like the Temporal Fusion Transformer and the Temporal Convolutional Network. However, they still need help with issues, specifically the higher computation cost and larger dataset for the training process. This research proposes a model that enhances energy forecasting for electric vehicles by employing outlier correction and a time series dense encoder as a forecasting technique. The model's novelty is the low computation cost, consideration of the data lagging, and day series past covariates. The model can be trained on limited data to maintain accuracy and effectiveness. Provides precise predictions of electric vehicle energy consumption based on data lagging, resulting in efficient resource management. This research significantly enhances the energy sector's efficiency and long-term viability by pioneering advanced forecasting models and methodologies. The performance evaluation, conducted using metrics like mean squared error and mean absolute deviation, unequivocally underscores the superior forecasting capabilities of the proposed model compared to the transformer-based models. © 2025 The Authors

关键词： Electric vehicle charging stations Energy consumption forecasting Smart cities Time series dense encoder

来源：评论

学校读者我要写书评

暂无评论

Prediction of Blood Oxygen Saturation by Physiological Variables Using Machine Learning

Prediction of Blood Oxygen Saturation by Physiological Var...

引用

international conference on advanced research in Technologies, information, Innovation and Sustainability, ARTIIS 2024

作者： Rovira, Ronald H. Gómez, Óscar W. Montaño, Manuel Bayas, Marcia M. Figueroa, Junior Andrade, Carlos Efrain TECED - Research Group Universidad Estatal Peninsula de Santa Elena La Libertad Ecuador Universidad Nacional de Colombia Manizales Colombia

ISBN: (纸本)9783031834318

In recent times, the integration of emerging technologies such as the Internet of Things and artificial intelligence have considerably transformed healthcare. As technology advances, the healthcare sector faces growing challenges due to ever-increasing demand and traditional invasive procedures. In this context, this article presents a model as an effective alternative to conventional oximetry especially useful in places where medical equipment is limited. The model utilizes deep learning to estimate blood oxygen levels through Long Short-Term Memory (LSTM) networks, which consist of two main layers. These layers include key indicators such as heart rate (HR), respiratory impedance (RESP), and pulse, witch were sampled 53 individuals for these indicators. The first is an LSTM layer with 100 cells and 41,600 parameters with a hyperbolic tangent activation function, designed to capture complex temporal patterns. The second linearly activated, dense layer with 101 parameters converts these representations into a clear and useful output. The results provide a mean absolute error of 1.68 and a root mean square error of 2.45, which shows that the model is quite accurate and suitable for applications that require continuous monitoring. In this work, a study was done to see how the main body functions are related. It found that heart rate and pulse are strongly connected, but not as much with respiratory rate and oxygen saturation. Also, used a Random Forest algorithm to determine the relationship between these factors and SpO2 values. The analysis revealed that heart rate and pulse are highly correlated, while the correlation with respiratory impedance and oxygen saturation is low. Finally, when implementing the model with TensorFlow Keras, we observed that it generalizes well and shows no signs of overfitting, which is a positive indication of its reliability. The results of this study not only demonstrate how emerging technology can improve the accuracy of health data, b

关键词： Hyperbolic functions

来源：评论

学校读者我要写书评

暂无评论

EEG Epileptic Seizure Detection Based on Fused Brain Functional Networks 2nd

EEG Epileptic Seizure Detection Based on Fused Brain Functio...

引用

2nd international conference on Applied Intelligence, ICAI 2024

作者： Zhou, Pengfei Pan, Yijie Zhang, Xun Shou, Jibin Guo, Jiayang Xu, Meiyan Gu, Peipei College of Software Engineering Zhengzhou University of Light Industry Zhengzhou450000 China Ningbo Institute of Digital Twin Eastern Institute of Technology Ningbo China Zhejiang Moorgen Group Co. Ltd. Zhejiang310000 China Department of Hematology School of Medicine Xiamen University Xiamen361000 China National Institute for Data Science in Health and Medicine Xiamen University Xiamen361000 China College of Computer Science Minnan Normal University Zhangzhou363000 China

ISBN: (纸本)9789819619061

Epilepsy, characterized by abnormal neuronal discharges, is a widespread neurological disorder. Traditional detection methods primarily focus on extracting time-frequency features from raw electroencephalogram (EEG) signals, often overlooking the complex interactions between brain regions. To address this limitation, we propose an automated approach for epileptic seizure detection. We first construct a fused brain network (TFPP) by integrating the Pearson Correlation Coefficient (PCC) and Phase Lag Index (PLV) across both temporal and frequency domains, enhancing EEG signals by capturing synchronized activities across different brain regions. Next, we introduce the CNN_sk_BiLSTM model, which leverages Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory (BiLSTM) networks to sequentially extract spatial and temporal features. The SKAttention mechanism further refines feature extraction across multiple scales. Using the CHB-MIT scalp EEG dataset, the proposed method achieved an average sensitivity of 90.86%, specificity of 98.79%, and accuracy of 97.88%. These results confirm that fused brain functional networks effectively integrate the intra-channel characteristics of EEG signals, demonstrating the superior performance of the proposed approach in epileptic seizure detection. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

1st international Symposium on Intelligent Technology for Future Transportation, ITFT 2024

1st International Symposium on Intelligent Technology for Fu...

引用

1st international Symposium on Intelligent Technology for Future Transportation, ITFT 2024

ISBN: (纸本)9783031841477

The proceedings contain 32 papers. The special focus in this conference is on Intelligent Technology for Future Transportation. The topics include: Modeling Autonomous Delivery Robots Under Framework of Automated Driving System Using Dynamic Transport Assignment: A State-of-the-Art;virtual Variable Baseline Stereo Vision Algorithm;A Conceptual Framework to Operational Design Domain (ODD)-Based Scenario Generation for Technical Evaluation Obstacle Detection in Automated Shuttle Bus;precise Driver’s Drowsiness Detection Using a Combination of Proven Methods with a Single Layer Neural Network;fusionSis: Analysis and Evaluation Framework for Fusion Safety State of Connected and Automated Vehicles Under Cyber Attacks;deep Learning-Based Monocular Depth Estimation Method for Forklift Collision Avoidance;enhancing Multi-user Experience: Optimizing Explanation Timing Through Game Theory;user-Intimate and Trustworthy Ontology-Based Requirement engineering Methodology: A Case Study on Connected and Autonomous Electrical Vehicles;AI and Leadership in Startup Innovation and Disruption: The Case of Mobility as a Service (MaaS);research on Traffic Analysis Zones Division Model for New Roads Based on Complex Network Theory;forecasting Vehicle Mobility on Various Segments of the Urban Transport Network in the Fuzzy Paradigm;The Emerging ICT for Exposure Reduction in Urban Mobility: Transport Models for Risk Cycle;concept of a Virtual Test Field for Inland Waterway Transport;critical Characterization of Three-Phase Traffic Flow in Severe Condition;design and Implementation of Traffic Congestion Relief Strategies Based on Multi-objective Optimization Algorithms;wide-Area Ship Movement Prediction Using Random Forests;characteristics of Heterogeneous Traffic Flow Involving Different Intelligent Level Autonomous Vehicles;the research on Customer Demand of Asia-Europe Liner Shipping Companies Based on Kano Model.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enhancing Domain Adaptation in Continual Learning with Elastic Weight Consolidation: A Multi-Dataset Deep Learning Approach 27th

Enhancing Domain Adaptation in Continual Learning with Elast...

引用

27th international conference on Pattern Recognition, ICPR 2024

作者： Premalatha, R. Jayanthi, K.B. Rajasekaran, C. Sureshkumar, R. Dhanalakshmi, R. K.S. Rangasamy College of Technology Namakkal India Erode Cancer Centre Erode India Indian Institute of Information Technology Tiruchirappalli India

ISBN: (纸本)9783031876592

Transfer learning (TL) is a powerful machine learning tool, particularly deep learning, enabling models to adapt to new tasks by leveraging prior knowledge. However, continual learning poses challenges, especially with domain adaptation. This work addresses these challenges by initially training a model from scratch on the CIFAR-100 dataset (source task), then transferring its knowledge to an intermediate task involving a Kaggle cancer dataset, followed by an additional transfer to a lung infection dataset (target task). While the models perform strongly on the intermediate task, they struggle on the target task due to continual learning challenges. To overcome this, the paper proposes using Elastic Weight Consolidation (EWC) to significantly improve test accuracy on the target task across various architectures, highlighting the importance of domain adaptation techniques in continual learning. Specifically, InceptionV3 demonstrates superior performance with notably improved metrics, achieving a test loss reduction of 0.82. EWC proves particularly effective when models are trained from scratch, as it helps to preserve important features by penalizing changes to critical weights. However, its effectiveness diminishes when applied to pre-trained models, such as those initialized with ImageNet weights, due to the optimized nature of these weights for general image recognition tasks. Thus, EWC is shown to be beneficial in scenarios involving models trained from scratch. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Stochastic Modeling of Path Length Dynamics in Ethereum’s Merkle Patricia Tries 19th

Stochastic Modeling of Path Length Dynamics in Ethereum’s M...

引用

19th international conference on Mathematical Modeling and Simulation of Systems, MODS 2024

作者： Kuznetsov, Oleksandr Kuznetsova, Kateryna Yezhov, Anton Chernushevych, Valentyn Zpoken OU Harju Maakond Tallinn Kesklinna linnaosa Sakala tn 7-2 Tallinn10141 Estonia Department of Theoretical and Applied Sciences eCampus University Via Isimbardi 10 CO Novedrate22060 Italy Department of Information and Communication Systems Security School of Computer Sciences V. N. Karazin Kharkiv National University 4 Svobody Sq. Kharkiv61022 Ukraine

ISBN: (纸本)9783031907340

This study presents a novel probabilistic framework for analyzing path length distributions in Merkle Patricia Tries, the foundational data structure of Ethereum’s state management system. We develop a mathematical model characterizing the probability distribution of path lengths for randomly generated Ethereum addresses and validate it through extensive computational simulations. Our findings reveal the logarithmic growth of average path lengths with respect to the number of addresses, confirming a critical property for Ethereum’s scalability. The study demonstrates remarkable accuracy in predicting average path lengths, with discrepancies between theoretical and empirical results remaining below 0.01 across tested scales from 1000 to 300 million addresses, encompassing Ethereum’s current state size. We identify and confirm the right-skewed nature of path length distributions, providing essential insights into worst-case scenarios and informing optimization strategies. Statistical analysis, including chi-square goodness-of-fit tests, strongly supports the model’s accuracy. This research bridges a significant gap between theoretical computer science and practical blockchain engineering, offering immediate applications for Ethereum client optimization, stateless client development, and informed protocol upgrade decisions. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Size distribution

来源：评论

学校读者我要写书评

暂无评论

Measuring Bit Error Rate in Plastic Fiber Optics Under Bending Conditions

Measuring Bit Error Rate in Plastic Fiber Optics Under Bendi...

引用

2025 IEEE SoutheastCon, SoutheastCon 2025

作者： Vidal, Alicia Carranza, Jade Nicotra, Philip Guerrero, Jorge Losada, Ángeles López, Alicia Gascón, Javier Mateo Carro, Pedro Ortega, Alfonso Lleida, Eduardo Ribas, Dayana Antoniades, Neophytos A. Richards, Dwight Jiang, Xin Wang, Yi College of Staten Island The City University of New York Engineering & Environmental Sciences Department New York10314 United States Aragón Institute of Engineering Research University of Zaragoza Electronic and Communications Engineering Department 50018 Spain Manhattan University Electrical and Computer Engineering Department New York10471 United States

ISBN: (纸本)9798331504847

Plastic optical fibers, POF, play a significant role in telecommunications. Their advantages make them a popular choice against their glass fiber counterparts and copper cables. With their resistant material and efficiency in data transmission, not to mention cost-effectiveness due to easier installation compared to that of glass fiber, applications of POF are surging. Despite being a popular choice, POF does have its limitations, mainly scarce power budget that can be impaired when transmitting under bending conditions. Investigation regarding POF behavior under bending conditions is needed to gain better understanding of the limits to their transmission properties. In this paper, additional information is provided regarding 3 fiber types: step-index, graded-index, and multicore step-index fibers, and their behaviors under bending conditions. The transmission properties measured are bit error rate, also known as BER, and optical power. This paper investigates the effects, as well as the relationship, between optical power and BER under these conditions. © 2025 IEEE.

关键词： Optical fiber communication

来源：评论

学校读者我要写书评

暂无评论

Enhancing Rhetorical Role Identification in Legal Documents Using Large Language Models and IN_place Data Augmentation 3rd

Enhancing Rhetorical Role Identification in Legal Documents...

引用

3rd international conference on Computing, Communication and Learning, CoCoLe 2024

作者： Sheik, Reshma Namani, Chandrababu Prasanna, P. Nirmala, S. Jaya National Institute of Technology Tamil Nadu Trichy India TKM College of Engineering Kerala Kollam India MVSR Engineering College Telangana Hyderabad India

ISBN: (纸本)9783031790409

Understanding the rhetorical roles of sentences within legal documents is crucial for various downstream tasks, including semantic search, summarization, and case law analysis. However, the complex structure of legal case documents, coupled with the interplay of various themes, poses challenges even for human experts. In this paper, we perform data augmentation using open-sourced Large Language Models (LLMs) to automate the identification of rhetorical roles. Specifically, we explore data augmentation techniques to address class imbalance issues within datasets, introducing a novel augmentation technique called IN_Place Augmentation to mitigate linear dependency issues. Additionally, a comparative analysis of two neural network architectures, InLegalBERT and Hierarchical BiLSTM, combined with Conditional Random Fields (Hier-BiLSTM-CRF), integrating different types of sentence embeddings was conducted. We fine-tune the InLegalBERT model using our proprietary dataset and utilize the fine-tuned embeddings to train the Hier-BiLSTM-CRF model. Our evaluations demonstrate the efficacy of the fine-tuned InLegalBERT model across diverse legal contexts, showcasing significant improvements in the Land & Property domain with an increase in the weighted average F1 score from 0.729 to 0.884, and in the Criminal domain with an increase in the macro average F1 score from 0.631 to 0.817. In summary, our contributions include leveraging LLM- Mistral 7B for data augmentation, introducing IN_Place augmentation, utilizing domain-specific transformer InLegalBERT, and developing a hybrid model integrating InLegalBERT with Hier-BiLSTM-CRF. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Leveraging Block-DAO Integration for Sustainable Smart Manufacturing: Operational Efficiency, Supply Chain Transparency, and Data Security Implications 2nd

Leveraging Block-DAO Integration for Sustainable Smart Manuf...

引用

2nd international conference on Sustainable Development Using Machine Learning, Artificial Intelligence, and IoT, ICSD 2024

作者： Dutta, Pushan Kumar Bhattacharya, Pronaya Sriram, K.P. Vijayakumar, K. Amity School of Engineering and Technology Amity University Kolkata700135 India Department of Information Technology St. Joseph’s Institute of Technology OMR Chennai600119 India

ISBN: (纸本)9783031717284

The shift to Industry 5.0 and sustainable smart manufacturing will require new governance and decision-making paradigms within digital manufacturing systems. To address this need, this work suggests that the incorporation of blockchain technology and Distributed Autonomous Organizations can be a disruptive approach to enhancing operational effectiveness, supply chain transparency, and data security in smart manufacturing. DAOs, made feasible by blockchain’s decentralized and transparent architecture, present a promising possibility for redefining cooperation and decision-making across stakeholders. The combination of the two generates a framework in which the inalterability and traceability of blockchain are used to guarantee supply chain transparency and data integrity, while the collective intelligence and democratic governance of DAOs are used to increase operational efficiency and sustainable practices. The proposed approach empowers stakeholders to participate in decision-making processes, fostering a collaborative ecosystem that optimizes resource allocation, minimizes waste, and promotes circular economy principles. By exploring real-world use cases and implementation strategies, this paper sheds light on the potential of block-DAO integration to address critical challenges in smart manufacturing, such as supply chain complexities, data silos, and environmental concerns. Additionally, it delves into the implications for data security and privacy, outlining robust mechanisms to safeguard sensitive information while maintaining transparency. Ultimately, this research paves the way for a sustainable and resilient smart manufacturing paradigm, where operational efficiency, supply chain transparency, and data security are seamlessly integrated through the synergistic combination of blockchain and DAO technologies. The proposed work provides a comprehensive guide for researchers, industry professionals, and policymakers seeking to leverage cutting-edge technologies

关键词： Smart manufacturing

来源：评论

学校读者我要写书评

暂无评论

Greenhouse Gas Prediction Using LSTM Algorithm Based on Microsensor in Bandung City, Indonesia 1st

Greenhouse Gas Prediction Using LSTM Algorithm Based on Micr...

引用

1st international conference on Machine Learning Algorithms, ICMLA 2024

作者： Adiwidya, Andre Suwardana Alexandra, Tania Christiana Kurniawan, Michelle Ummi, Annisa Zahwatul Halinda, Maulana Fauzan Athalla Al Gyfary Oktaviany, Indah Cikal Kampong, Prichel Adisatya Lalintia, Irvin Judah Lee, Vivian Amalia, Dini Rizqi Putra, Nabilah Indira Rohma, Lailatul Salam, Rahmat Awaludin Chandra, Indra School of Electrical Engineering Telkom University Bandung Indonesia Engineering Physics School of Electrical Engineering Center of Excellence of Sustainable Energy and Climate Change Telkom University Bandung Indonesia

ISBN: (纸本)9783031758607

Greenhouse gases such as Carbon Dioxide (CO2), Methane (CH4), and Ozone (O3) threaten human health and the environment by polluting the air. Previous research at Telkom University, Bandung, used the Backpropagation Artificial Neural Network (ANN) method to predict Particulate Matter 2.5 (PM2.5) concentrations. The results show Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) values of 8.32 μg/m3 and 37% at the GKU measurement station (a height of 35 m) and 12.49 μg/m3 and 15% at the Deli measurement station (height 15 m). The CO2, CH4, and O3 prediction parameters were combined at the Telkom University Landmark Tower (TULT) measuring station (70 m height) to optimize the prediction model using deep learning techniques to improve the system. The prediction model is assessed using the Long Short-Term Memory (LSTM) algorithm, which stores relevant information for long periods. This reduces RMSE values: 0.089923 for CO2, 0.060467 for CH4, and 0.036242 for O3. This improved LSTM model can estimate measured gas levels for the next hour, so it has the potential to be applied as an early warning system for air quality in Bandung as well as to provide information about GHG exposure to the public. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Mean square error

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：