检索结果-内蒙古大学图书馆

17th International Conference on COMmunication Systems and NETworkS, COMSNETS 2025

作者： Banik, Swarnali Adithya, Akhilesh Mitra, Bivas Sen, Sougata Saha, Snehanshu Ghosh, Surjya Amazon India Indian Institute of Technology Department of Computer Science and Engineering Kharagpur India BITS Pilani K K Birla Goa Campus Department of Computer Science and Information Systems India BITS Pilani K K Birla Goa Campus APPCAIR Goa India

ISBN: (纸本)9798331531195

We present an annotation tool designed to collect fine-grain continuous emotion ground truth labels and physiological responses as users watch different videos. In the existing literature, continuous emotion annotations are typically collected as self-reports based on the 2D Circumplex model through auxiliary devices (a joystick or similar tools) during video consumption. However, the users face challenges as they need to use an auxiliary device and rely on the Circumplex model's valence and arousal axis without clear indication of distinct emotion labels. Therefore, to overcome these challenges, we introduce PResCon, a tool that allows to collect emotion annotations using keyboard arrow keys only (without using any auxiliary device) and guiding users towards the discrete emotions (e.g., amused, stressed, bored, relaxed) by embedding them on the Circumplex plane. The tool encompasses two major components - (a) a web interface that displays various stimuli videos and provision of collecting emotion labels (valence and arousal) by moving the keyboard arrow keys and (b) a sensor setup (based on Arduino) that allows to collect the physiological responses (heart rate and galvanic skin response) passively during video consumption. In summary, the PResCon tool facilitates research in the domain of affective computing by allowing to collect fine-grain continuous emotion annotations and the physiological responses in real-time during video consumption. © 2025 IEEE.

关键词： Auxiliary equipment

来源：评论

学校读者我要写书评

暂无评论

An Overview of Machine Learning Applications in Blockchain for Green Energy Exchanges 33

An Overview of Machine Learning Applications in Blockchain f...

引用

33rd Southern African Universities Power engineering Conference, SAUPEC 2025

作者： Nemakonde, Pfano Nemangwele, Fhulufhelo Ratshitanga, Mukovhe University of Venda Faculty Science Engineering and Agriculture Dept. of Physics Thohoyandou South Africa Cape Peninsula University of Technology Dept. of Electrical Electronic and Computer Engineering Bellville South Africa

ISBN: (纸本)9798331535162

There is an increasing demand for affordable, decentralised and distributed electricity, which is one of the key motivating factors that has incentivised this review article's writing. Industrial 4.0 has been a conduit through which the conventional grids shift towards a self-autonomous system, and blockchain platforms are emerging as viable solutions to facilitate that paradigm. With the advent of peer-to-peer (P2P) energy trading, community-based energy exchanges have significantly contributed to reduced grid dependency. Machine learning (ML) is being implemented in predictive analytics for energy forecasting (EF), thereby aiding both distribution network operators (DNOs) and aggregators to precisely and predictably monitor, optimise and stabilise grid operations through smart agents. The main aim of this review paper is to provide a detailed analysis of the existing research and implementation approaches on blockchain-based P2P energy trading and a comparative overview of how ML applications aid predictive analytics in community-based energy exchanges. © 2025 IEEE.

关键词： Predictive analytics

来源：评论

学校读者我要写书评

暂无评论

Enhanced Facial Emotion Detection Using Convolutional Neural Networks and Attention Mechanisms 5

Enhanced Facial Emotion Detection Using Convolutional Neural...

引用

5th IEEE International Conference on Advances in electrical, Computing, Communication and Sustainable Technologies, ICAECT 2025

作者： Indhumathy, D. Yeswanth, S. Sri Ramakrishna Engineering College Department of Mathematics Coimbatore India Bannari Amman Institute of Technology Department of Computer Science and Engineering Sathyamangalam Erode India

ISBN: (数字)9798350356236

ISBN: (纸本)9798350356236

Convolutional neural networks are now facing a number of technological obstacles in the field of facial expression identification, including the low automatic recognition rate, the difficulty of effectively recognizing minor changes in facial emotions, and the complexity of feature extraction. In order to enhance automatic facial expression identification, this research suggests a Height Performance Module Implement (HPMI) attention mechanism based on spatial attention. The model concentrates on important spatial characteristics by incorporating this attention mechanism, which improves recognition during training. The VGG-16 network incorporates a spatial attention-based HPMI module that efficiently reduces over fitting, amplifies key information, and suppresses irrelevant data. This facilitates effective information exchange between the network model and the image's key properties. It also deals with the problem of dimensional mismatch between output and input. On the CK+ expression datasets, the approach yields accuracy rates of 97.05%, respectively. According to experimental results, the incorporation of the HPMI module improved accuracy by an average of 3.94%. © 2025 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Enhanced satellite imagery analysis for post-disaster building damage assessment using integrated ResNet-U-Net model

引用

Multimedia Tools and Applications 2025年第5期84卷 2689-2714页

作者： Bhardwaj, Diwakar Nagabhooshanam, N. Singh, Ajeet Selvalakshmi, B. Angadi, Sanjeevkumar Shargunam, S. Guha, Tapas Singh, Gurkirpal Rajaram, A. Computer Engineering and Applications GLA University Chaumuhan India Department of Research Analytics Saveetha Dental College and Hospitals Saveetha Institute of Medical and Technical Sciences Saveetha University Chennai India Aditya University Surampalem Andhra Pradesh India VIT Bhopal University Madhya Pradesh Bhopal466114 India Department of Computing Science and Engineering Tagore Engineering College Rathinamangalam Tamil Nadu Chennai600127 India Department of Computing Science and Engineering Nutan College of Engineering and Research Talegaon Dabhade Maharashtra 410507 India Department of Computer Science and Engineering Kalasalingam Academy of Research and Education Tamil Nadu Srivilliputhur626126 India Department of CSE Techno International New Town West Bengal Kolkata700156 India Department of Architecture PTU Punjab Mohali140307 India Department of Electronics and Communication Engineering E.G.S. Pillay Engineering College Tamil Nadu Nagapattinam611002 India

In the aftermath of natural disasters, timely and accurate building damage assessment is crucial for effective disaster response. This research introduces an innovative integrated ResNet-U-Net ensemble model that significantly enhances the accuracy and efficiency of damage detection from satellite imagery. The model combines the deep feature extraction capabilities of ResNet with the precise localization advantages of U-Net, supplemented by attention mechanisms to focus on critical damage indicators. Methodologically, the research employs a dataset of high-resolution satellite images from post-Hurricane Iota, processed through normalization and augmentation techniques. The model demonstrates exceptional accuracy, achieving 96.91% on an unbalanced dataset representative of real-world conditions, and 89.35% on a balanced dataset. This performance indicates a substantial improvement over existing approaches, offering a rapid and reliable tool for aiding disaster relief efforts. The proposed model's computational efficiency and generalizability underscore its potential to transform disaster management practices. Future work will explore the model's application to various disaster types and integration with real-time data processing systems. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Satellite imagery

来源：评论

学校读者我要写书评

暂无评论

An Approach for Detecting and classification of Lung Cancer Using Deep Learning and Machine Learning 14

An Approach for Detecting and classification of Lung Cancer ...

引用

14th IEEE International Conference on Communication Systems and Network Technologies, CSNT 2025

作者： Sonal, Deeksha Pattanaik, Srinibas Nayak, Kanishk Vyas, Rudraksh Raj Sharma, Tanya Kohli, Ayush Department of CSE Computer Science & Engineering Chandigarh University Punjab India Department of CSE UIE University Institute of Engineering Chandigarh University Punjab India

ISBN: (纸本)9798331531935

Lung cancer is one of the top causes of cancer-related fatalities worldwide, demanding early and accurate detection for better patient outcomes. This work addresses the merging of neural network (ML) along with deep learning (DL) approaches to increase lung cancer diagnosis and classification. Using sophisticated convolution neural networks (CNNs) for image analysis and standard machine learning (ML) models like random forest algorithms and supported vector machines (SVMs) for structured data, the suggested technique achieves excellent diagnostic accuracy, avoiding false positives and negatives. Enhanced CNNs give higher efficiency with precision of 93%, recall of 98.5%, and an accuracy of 97%. Key indicators include cumulative risk, age, and weariness, whereas demographic characteristics have a modest influence. Visualization methods such as feature significance charts and confusion matrices verify model dependability. This hybrid AI-driven approach indicates great promise in transforming lung cancer diagnoses, however problems including data privacy, computing needs, and validation in clinical trials remain and This work utilizes artificial intelligence (AI) (ML) in conjunction with deep learning (DL) technology to produce a more precise and automated approach for lung cancer diagnosis and categorization. © 2025 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Execution Time Optimization for Pipeline Deep Network Training on Multiple GPUs 33

Execution Time Optimization for Pipeline Deep Network Traini...

引用

33rd Euromicro International Conference on Parallel, Distributed, and Network-Based Processing, PDP 2025

作者： Wu, Bing-Jou Hong, Ding-Yong Liu, Pangfeng Wu, Jan-Jan National Taiwan University Department of Computer Science and Information Engineering Taipei Taiwan Institute of Information Science Academia Sinica Taipei Taiwan

ISBN: (纸本)9798331524937

As neural network models become gigantic, they increasingly demand more time and memory for training. To meet these demands, advanced parallel computing techniques have become essential. Our research focuses on hybrid parallelism, an extension of pipeline parallelism. Pipeline parallelism splits the neural network into sub-networks distributed across a sequence of processing units, enabling simultaneous processing of different data segments on each device. Hybrid parallelism extends this concept by allocating multiple devices to each sub-network. Our research focuses on optimizing hybrid parallelism by improving how the model is partitioned and how computational devices are assigned. We address these issues by modeling the neural network as a directed acyclic graph of tensor operators, and then demonstrating that optimally partitioning this graph is NP-complete. Then, we propose a two-step approach. The first step is to determine a sequence of nodes. The second step is dynamic programming, which partitions the sequence to maintain balance across the assigned devices. In transforming the graph into a sequence, we explore two methods: one employs topological sorting, while the other clusters non-sequential subgraphs. We apply both methods and select the more effective one based on performance outcomes. We implement our algorithm and conduct experiments. The results show substantial enhancements in both the speed of partitioning and training throughput, with speedups reaching up to 23 in partitioning time and a 1.3 -fold increase in training throughput. © 2025 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Single Microphone Speech Denoising Using Wavelet Thresholding and RLS Algorithm

Single Microphone Speech Denoising Using Wavelet Thresholdin...

引用

International Conference on Electronics, Information and Communications (ICEIC)

作者： Yashar Naderahmadian Seyed Ghorshi Issa Panahi Department of Electrical and Computer Engineering University of Guilan Rasht Iran Department of Electrical and Computer Engineering The University of Texas at Tyler Tyler Texas USA

ISBN: (数字)9798331510756

ISBN: (纸本)9798331510763

This work demonstrates a practical and effective solution for speech denoising with reduced hardware requirements. This paper proposes a novel speech denoising method that combines wavelet denoising with Recursive Least Squares (RLS) filtering, requiring only one microphone. The traditional RLS algorithm relies on two microphones to suppress noise: one for noisy speech and another for noise samples. Our approach improves upon this by using wavelet denoising to provide an initial noise estimate, which is then refined using RLS filtering. We evaluate the denoising performance of the proposed method against the traditional RLS algorithm and standalone wavelet denoising, using Mean Squared Error (MSE) and Perceptual Evaluation of Speech Quality (PESQ) as the performance metrics. The evaluation, conducted on speech samples from the TIMIT voice corpus under various noise conditions and signal-to-noise ratios, shows that our method consistently outperforms the traditional RLS and wavelet-only approaches. Spectrogram analyses further confirm the superiority of our method in noise suppression while preserving the integrity of the original signal.

关键词： Noise reduction Transforms Speech enhancement Information filters Hardware Noise measurement Spectrogram Signal integrity Microphones Signal to noise ratio

来源：评论

学校读者我要写书评

暂无评论

PlantRx: Harnessing Deep Learning for Medicinal Plant Identification 6th

PlantRx: Harnessing Deep Learning for Medicinal Plant Identi...

引用

6th International Conference on Smart Computing and Informatics, SCI 2024

作者： Yacoob, Shaik Sindhu, B. Kavya, V. Nagalakshmi, A. Chandra Sekhar Reddy, N. Department of Computer Science and Engineering Godavari Institute of Engineering and Technology Andhra Pradesh Rajahmundry India Department of Computer Science and Engineering Dr. M.G.R Educational and Research Institute Maduravoyal Tamil Nadu Chennai India

ISBN: (纸本)9789819619801

The real-time identification of medicinal plants holds substantial significance in various domains such as ethno botany, pharmaceuticals, and traditional medicine. This paper proposes an innovative approach that leverages deep learning techniques to achieve swift and accurate identification of medicinal plant species. Traditional methods of plant identification are often time-consuming and require specialized botanical expertise. In contrast, our proposed DL-based system harnesses the power of computer vision and pattern recognition to instantly recognize medicinal plants directly from visual input, such as images of leaves, flowers, or whole plants. It employs deep learning architectures that enable the model to automatically learn distinctive features and patterns unique to each plant species. This research introduces an intelligent vision-based system designed to recognize herb plants through the creation of an automated Convolutional Neural Network (CNN). The articulated deep learning model incorporates a CNN block dedicated to the extraction of relevant features. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Convolutional neural networks (CNN) Deep learning Medicinal Plants PlantRx Species identification

来源：评论

学校读者我要写书评

暂无评论

Improving graph collaborative filtering with network motifs

引用

Neural Computing and Applications 2025年 1-20页

作者： Zhang, Yuqi Yu, Jian Liu, Zhizhong Wang, Guiling Nguyen, Minh Sheng, Quan Z. Wang, Nancy Department of Computer and Information Sciences Auckland University of Technology Auckland Central1010 New Zealand School of Computer and Control Engineering Yantai University Shandong Yantai264005 China School of Information Science and Technology North China University of Technology Beijing100144 China School of Computing Macquarie University SydneyNSW2109 Australia

Deep learning on graphs, specifically graph convolutional networks (GCNs), has exhibited exceptional efficacy in the domain of recommender systems. Most GCNs have a message-passing architecture that enables nodes to aggregate information from neighbours iteratively through multiple layers. This enables GCNs to learn from higher-order information, but the model does not allow for direct captions of the local structural patterns. Our rationale is to investigate the effectiveness of capturing such local patterns for graph-based collaborative filtering to enhance model’s learning ability per layer. This technique combines lower-order and higher-order interactions during layer-wise propagation. In this paper, we propose MotifGCN to aggregate both lower-order and higher-order information in each graph convolution layer. Specifically, we develop dedicated algorithms of generating motif adjacency matrices. The matrices are then used for motif-enhanced neighbourhood aggregation in each layer. As this paper focuses on recommender systems, MotifGCN is built on the basis of bipartite graphs. Our experiments on four real-world datasets show that MotifGCN has a superior performance compared to various state-of-the-art methods. © The Author(s) 2025.

关键词： Collaborative filtering

来源：评论

学校读者我要写书评

暂无评论

Exploring privacy mechanisms and metrics in federated learning

引用

Artificial Intelligence Review 2025年第8期58卷 1-51页

作者： Shenoy, Dhanya Bhat, Radhakrishna Krishna Prakasha, K. Department of Computer Science and Engineering Manipal Institute of Technology Manipal Academy of Higher Education Karnataka Manipal576104 India Department of Information and Communication Technology Manipal Institute of Technology Manipal Academy of Higher Education Karnataka Manipal576104 India

The federated learning (FL) principle ensures multiple clients jointly develop a machine learning model without exchanging their local data. Various government enacted prohibition policies on data exchange between organizations have led to the need for privacy-preserved federated learning. Many industries have cultivated this idea of model development through federated learning to enhance performance and accuracy. This paper offers a detailed overview of the background of FL, highlighting existing aggregation algorithms, frameworks, implementation aspects, and dataset repositories, establishing itself as an essential reference for researchers in the field. The paper thoroughly reviews existing centralized and decentralized FL approaches proposed in the literature and gives an overview about the methodology, privacy techniques implemented and limitations to guide other researchers to advance their research in the field of federated learning. The paper discusses the critical role of privacy-enhancing technologies like differential privacy (DP), homomorphic encryption (HE), and secure multiparty computation (SMPC) in federated learning highlighting their effectiveness in safeguarding sensitive data while optimizing the balance between privacy, communication efficiency, and computational cost. The paper explores the applications of federated learning in privacy-sensitive areas like natural language processing (NLP), healthcare, and Internet of Things (IoT) with edge computing. We believe our work provides a novel addition by identifying privacy evaluation metrics and spotlighting the measures in terms of data privacy and correctness, communication cost, computational cost and scalability. Furthermore, it identifies emerging challenges and suggests promising research directions in the federated learning domain. © The Author(s) 2025.

关键词： Federated learning

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：