检索结果-内蒙古大学图书馆

24th International Conference on Algorithms and Architectures for Parallel Processing, ICA3PP 2024

作者： Xu, Nan Yang, Baisong Deng, Lihao Ren, Jia The Institute of Data Science City University of Macau Macau China The College of Information and Communication Engineering Hainan University Haikou China

ISBN: (纸本)9789819615414

Tropical Cyclone tracks serve as crucial criteria for discerning the affected regions and the extent of impact caused by tropical cyclones. Classifying tropical cyclone tracks allows for the exploration of tropical cyclone characteristics and the assessment of the associated impacts. This paper utilized the RSMC-Tokyo dataset provided by the Japan Meteorological Agency (JMA), spanning the years 2000 to 2021. Employing the K-Means clustering algorithm, 13 variables describing both the intrinsic features of tropical cyclone tracks and their external impacts were selected as clustering indicators. The tropical cyclones in the northwestern Pacific region were classified into four categories, and a comprehensive feature analysis was conducted for each class. Results indicate significant differences in the activity range, intensity, seasonal occurrence, and landfall patterns among the identified tropical cyclone classes. Notably, tropical cyclones originating in the South China Sea exhibit characteristics such as rapid development, high intensity, unpredictability, short cycles, and a greater propensity to affect Hainan Province. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Tropical cyclone

来源：评论

学校读者我要写书评

暂无评论

Optimization-Inspired Graph Neural Network for Cellular Network Optimization

引用

IEEE Transactions on Mobile Computing 2025年第7期24卷 6446-6459页

作者： He, Pengcheng Tang, Yijia Xu, Fan Shi, Qingjiang Tongji University School of Computer Science and Technology Shanghai China Shenzhen Research Institute of Big Data Shenzhen China Tongji University School of Software Engineering Shanghai China Tongji University School of Electronic and Information Engineering Shanghai China

The rapid development of wireless communications have driven the need for careful optimization of network parameters to improve network performance and reduce operational cost. Traditional methods, however, struggle with the vast number of tunable parameters and lack scalability in diverse network scenarios. To address these challenges, this paper introduces an optimization-inspired bipartite graph neural network (Bi-GNN) approach for scalable network optimization. Our approach leverages the bipartite structure of network topologies, and incorporates a message-passing mechanism by unfolding the Zeroth-Order Block Coordinate Projected Gradient Descent (ZO-BCPGD) algorithm, which ensures not only high-performance optimization but also manageable computational demand. We demonstrate the permutation and dimensionality equivariance property of the Bi-GNN, which significantly enhances the model's generalizability across various network structures and sizes. Furthermore, we theoretically analyze the expressive power and generalization ability of the Bi-GNN, demonstrating its adeptness at complex network optimization tasks. The training process, parallel execution, and practical implementation techniques are also discussed to ensure the model's applicability in real-world scenarios. Numerical results verify that the Bi-GNN outperforms existing methods in both coverage ratios and computational cost. Furthermore, our approach exhibits robust scalability across various network scenarios, making it a versatile tool for optimizing a wide range of wireless networks. © 2002-2012 IEEE.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Effortless learning: leveraging summarization and tabulation for automated note making

引用

Multimedia Tools and Applications 2024年 1-26页

作者： Arivarasi, A. Srinivasan, R. Rajeswari, D. Govindasamy, Alagiri School of Electronics Engineering Vellore Institute of Technology Tamil Nadu Chennai600127 India Department of Computing Technologies School of Computing College of Engineering and Technology SRM Institute of Science and Technology Kattankulathur603203 India Department of Data Science and Business Systems School of Computing College of Engineering and Technology SRM Institute of Science and Technology Kattankulathur603203 India Future Connect Technologies Pvt. Ltd Tamil Nadu Kanchipuram600034 India

In this fast-paced Digital Age, Natural Language Processing (NLP) can prove beneficial in consuming quality information efficiently. With the ever-growing number of learning resources, it is becoming onerous for students to choose the best-suited learning resource. Furthermore, many students find it difficult to understand and retain highly factual information. To overcome these issues, this paper proposes a novel Automated TExt Summarization of scholastic text in Table format (Auto-TEST) to provide information at a glance by mimicking the student’s style of note-making. The Auto-TEST framework is achieved through two steps which is Summarization and Tabulation. Initially, the Extractive Single Document Text Summarization (ESDTS) method is adopted to identify the salient information from the non-fictional (scholastic) text which is extracted and grouped together to form a concise summary. A Named Entity Recognition (NER) method is implemented to extract noun chunks and then arrange them in a tabulation format that aids in memorizing essential facts and figures by enabling effortless retrieval. The performance of the Auto-TEST framework is gauged by using scholastic domain-specific dataset using ROUGE as an evaluation metric. The extracted entities gives an average Recall of 71.40%, an average Precision score of 75.80% and an average F-score of 73.54% for the dataset respectively. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Signs to Speech 2

Signs to Speech

引用

2nd IEEE International Conference on Networking and Communications, ICNWC 2024

作者： Vardhan, Harsh Harsavardhan, S. Lakshmi, M. SRM Institute of Science and Technology School of Computing College of Engineering and Technology Department of Data Science and Business Systems Kattankulathur Chennai603203 India

ISBN: (纸本)9798350365269

This research introduces an innovative Sign Language to Speech Conversion Model using Convolutional Neural Networks (CNNs) to address communication barriers for the people who are deaf and unable to hear properly. The model employs deep learning for the automatic extraction of spatial patterns from sign language images. A diverse sign language dataset is carefully curated, undergoing Pre-processing for improved quality. The CNN architecture, designed for adaptability, captures local and global features, enabling accurate *** the training phase, the model learns to map sign language gestures to spoken language representations. Performance evaluation methods which include accuracy, recall, precision and F1 score, demonstrate the effectiveness of the model. Results underscore the significance for the deaf community. The discussion explores broader implications in education, healthcare, and social interactions, acknowledging limitations and proposing future research *** conclusion, the Sign Language to Speech Conversion Model contributes to sign language recognition, offering an advanced solution for communication barriers and promoting inclusivity in diverse social settings. © 2024 IEEE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Learning in sinusoidal spaces with physics-informed neural networks

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第3期5卷 985-1000页

作者： Wong, Jian Cheng Ooi, Chin Chun Gupta, Abhishek Ong, Yew Soon Institute of High Performance Computing Agency for Science Technology and Research Singapore138632 Singapore School of Computer Science and Engineering Nanyang Technological University Singapore639798 Singapore Singapore Institute of Manufacturing Technology Agency for Science Technology and Research Singapore138632 Singapore Agency for Science Technology and Research Singapore138632 Singapore Data Science and Artificial Intelligence Research Centre School of Computer Science and Engineering Nanyang Technological University Singapore138632 Singapore

A physics-informed neural network (PINN) uses physics-Augmented loss functions, e.g., incorporating the residual term from governing partial differential equations (PDEs), to ensure its output is consistent with fundamental physics laws. However, it turns out to be difficult to train an accurate PINN model for many problems in practice. In this article, we present a novel perspective of the merits of learning in sinusoidal spaces with PINNs. By analyzing behavior at model initialization, we first show that a PINN of increasing expressiveness induces an initial bias around flat output functions. Notably, this initial solution can be very close to satisfying many physics PDEs, i.e., falling into a localminimum of the PINN loss that onlyminimizes PDE residuals, while still being far from the true solution that jointly minimizes PDE residuals and the initial and/or boundary conditions. It is difficult for gradient descent optimization to escape from such a local minimum trap, often causing the training to stall. We then prove that the sinusoidalmapping of inputs-in an architecture we label as sf-PINN-is effective to increase input gradient variability, thus avoiding being trapped in such deceptive local minimum. The level of variability can be effectively modulated to match high-frequency patterns in the problem at hand. A key facet of this article is the comprehensive empirical study that demonstrates the efficacy of learning in sinusoidal spaces with PINNs for a wide range of forward and inversemodeling problems spanning multiple physics domains. Impact Statement-Falling under the emerging field of physicsinformed machine learning, PINN models have tremendous potential as a unifying AI framework for assimilating physics theory and measurement data. However, they remain infeasible for broad science and engineering applications due to computational cost and training challenges, especially for more complex problems. Instead of focusing on empirical demonstration of appli

关键词： Gradient methods

来源：评论

学校读者我要写书评

暂无评论

Forecasting Air Pollution of Urban Chennai Localities using Artificial Intelligence 3

Forecasting Air Pollution of Urban Chennai Localities using ...

引用

3rd International Conference on Applied Artificial Intelligence and Computing, ICAAIC 2024

作者： Faiz, Rayaan Agarwal, Tisha Rajeswari, D. Srm Institute of Science and Technology School of Computing College of Engineering and Technology Dept. of Data Science and Business Systems Chennai603203 India

ISBN: (纸本)9798350375190

India's air quality deteriorated significantly in 2023, ranking third worst globally, highlighting the urgency for effective monitoring and mitigation measures. To comprehend past trends and predict future patterns, historical air pollution data from seven urban areas in Chennai was analyzed using the XGBoost Algorithm. This study aims to gain insights into Chennai's air quality, identify abnormalities, and issue timely warnings to the public and governing authorities. XGBoost was employed to forecast AQI, NO2, PM2.5, NOx, PM10, CO, SO2, NH3, and Ozone levels, outperforming traditional LSTM, Prophet, and ARIMA models. Evaluation using RMSE and MAE scores revealed XGBoost's efficacy, with scores of 20.65 and 14.35, respectively. The research highlights regions like Alandur, Arumbakkam, and Manali with notable concentrations of contaminants, emphasizing the need for immediate action. These findings highlight the significance of utilizing advanced algorithms for accurate air quality forecasting and proactive pollution management. © 2024 IEEE.

关键词： Air quality

来源：评论

学校读者我要写书评

暂无评论

Violence Detection System using MobileNetV2 3

Violence Detection System using MobileNetV2

引用

3rd International Conference on Applied Artificial Intelligence and Computing, ICAAIC 2024

作者： Kumar, Rahul Gupta, Aniket Rajeswari, D. School of Computing College of Engineering and Technology Srm Institute of Science and Technology Department of Data Science and Business Systems Kattankulathur Chennai603203 India

ISBN: (纸本)9798350375190

The number of cases of violence and fights has been increasing around the world. With the use of CCTV, such incidents can be recorded but the detection of Violence is a major issue around the globe, and it plays a vital role in the safety and security of the area. It helps in effective surveillance and law enforcement. The effectiveness of any violence detection system depends on the speed and accuracy with which it detects the events and how it performs over different types of videos of varied sizes and formats. In the proposed algorithm, the dataset used contains videos of different types and scenarios to help the model understand and train better. To improve the overall effectiveness of the model, Lightweight CNN MobileNetV2 has been used along with that Bidirectional LSTM (BiLSTM) has been used. The accuracy rate of the proposed model is 98%. © 2024 IEEE.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Enhanced Heart Disease Diagnosis in MRI Images Using Swan Adapt Optimized Thresholding Segmentation with U-Net Model 5

Enhanced Heart Disease Diagnosis in MRI Images Using Swan Ad...

引用

5th International Conference on Electronics and Sustainable Communication Systems, ICESC 2024

作者： Steephan Amalraj, J. Kumaresan, T. Pravin Savaridass, M. Vaanathi, S. Bannari Amman Institute of Technology Computer Science and Engineering Sathyamangalam India Bannari Amman Institute of Technology Electronics and Instrumentation Engineering Sathyamangalam India Bannari Amman Institute of Technology Atrificial Intelligence and Data Science Sathyamangalam India

ISBN: (数字)9798350379945

ISBN: (纸本)9798350379945

The diagnosis of heart disease in MRI images is a critical application within the area of medical imaging. Conventional segmentation algorithms have difficulty in the accurate delineation of normal and sick heart tissues due to problems such as noise sensitivity, inaccuracy in boundaries, and sensitivity to parameters. In this work, we propose a novel strategy of thresholding based on image clusters, which we term SwanAdaptOpti thresholding for the segmentation of MRI images. The proposed system also utilizes Watershed segmentation with markers refined to enhance the segmentation of diseased heart tissues. Indeed, our approach is capable of achieving satisfactory performance in delineating normal and sick heart images, avoiding issues like overfitting, underfitting, and local optima. The overall robustness of our approach is demonstrated by extensive experimentation, which reveals the potential of our approach for the accurate and reliable diagnosis of heart disease from MRI images with an accuracy of 96%. © 2024 IEEE.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Computer vision and Generative AI for yield prediction in Digital Agriculture 2

Computer vision and Generative AI for yield prediction in Di...

引用

2nd IEEE International Conference on Networking and Communications, ICNWC 2024

作者： Majumder, Sayan Khandelwal, Yash Sornalakshmi, K. SRM Institute of Science and Technology School of Computing Faculty of Engineering and Technology Department of Data Science and Business Systems Kattankulathur Chennai603203 India

ISBN: (纸本)9798350365269

For the cause of evolution of agriculture to its next generation, the introduction of A.I. and data-driven approach is going to be an important part of the agricultural industry that as per our vision would offer numerous economic, environmental and social benefits. Controlled Environment Agriculture (CEA) is providing more benefits since the weather and other environmental conditions is controlled and predictable. In CEA, the benefits include enhanced productivity in yield, reduced environmental footprints and better resource management. Our solution uses the adoption of Generative AI to visually forecast the potential yield of a crop using stable diffusion and conditional prompts additionally implementing several Generative Adversarial Networks (GAN) architectures for data generation, augmentation and super-resolution. Our proposed system also adopts generative AI for real time monitoring of plants. studying their respective conditions and their autonomous cultivation and harvesting patterns. The dataset used contained over 8479 examples for different fruits alongside annotations. This data was used to train the detection models as well categorically feed the GANs for data generation. data generated by DCGAN has a structural similarity of 0.5-0.7 whereas data generated by pix2pix network returns us a structural similarity of 0.7-0.9. © 2024 IEEE.

关键词： Cultivation

来源：评论

学校读者我要写书评

暂无评论

Fill in the Gap: Two-Stage Prediction Modeling Framework for Understanding Human Biosensor data 15

Fill in the Gap: Two-Stage Prediction Modeling Framework for...

引用

15th International Conference on Information and Communication Technology Convergence, ICTC 2024

作者： Ro, Juneyoung Yoon, Yoonjin Graduate School of Data Science Korea Advanced Institute of Science and Technology Daejeon Korea Republic of Korea Advanced Institute of Science and Technology Department of Civil and Environmental Engineering Daejeon Korea Republic of

ISBN: (纸本)9798350364637

Human behaviors can be understood from various forms of information such as sleep, emotions or health conditions. With the aid of advanced technology, sensors can capture the conditions of humans and provide valuable insights. In this work, we used imputation strategies to overcome the challenge of incomplete data records and feature misalignment. A two-stage modeling of a Random Forest classifier was done to make final predictions on the test set. This relatively simple yet heavily data-centric approach leverages the strengths of multi-output classification and sequential learning by enhancing data integrity and feature consistency. © 2024 IEEE.

关键词： Prediction models

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：