检索结果-内蒙古大学图书馆

Artificial Intelligence in Maritime Anomaly Detection: A Decadal Bibliometric Analysis (2014–2024)

Journal of The Institution of Engineers (India): Series C 2025年第2期106卷 665-689页

作者： Thakur, Aman Singh Alex, T. Lawrence Nighojkar, Amrita Department of Mechanical Engineering National Institute of Technology Raipur Raipur India R & D Management and AI Analytics Research Group Defence Institute of Advanced Technology Pune India

This investigation builds upon scholarly research on AI-integrated anomaly detection in maritime safety operations from 2014 to 2024. The bibliometric analysis analyzes key Countries, Journals, Research themes, Keyword frequency, Co-citation analysis, Impactful authors, and significant collaborations among researchers and relevant institutes for Maritime Anomaly Detection Research. Our findings highlight substantial contributors, including China, France, Italy, and the USA, emphasizing essential collaboration networks and influential journals such as IEEE Transactions on Intelligent Transportation Systems and the international Journal of Approximate Reasoning. We scrutinize the application of various AI and machine learning models, such as LSTM, CNN, and SVM, across different anomaly types. These models are frequently implemented to detect environmental, navigation, cybersecurity, sensor, operational, communication, and behavioural anomalies. For navigation anomalies like unexpected course changes and speed variations, LSTM is utilized due to its capability to handle sequential data. CNN is employed for spatial-temporal pattern recognition in navigation routes, while SVM is applied to classify navigation patterns and detect deviations. Similarly, LSTM effectively identifies security anomalies, including unauthorized boardings and suspicious activities, through sequence modelling, and CNN proves helpful for spatial analysis. This study provides comprehensive insights into the current state of maritime anomaly detection research. It proposes future research areas, emphasizing model accuracy and efficiency enhancements, developing hybrid models, and addressing emerging challenges. Our analysis aims to guide researchers and practitioners by identifying critical areas for advancement and fostering potential collaborative efforts in this field. © The Institution of Engineers (India) 2025.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Applying machine learning to Improve Weather Forecasting in a Mediterranean Area: Preliminary Experiments 4th

Applying Machine Learning to Improve Weather Forecasting in...

引用

4th international conference on Numerical Computations: Theory and Algorithms, NUMTA 2023

作者： Yousaf, Umair De Rango, Alessio Furnari, Luca D’Ambrosio, Donato Senatore, Alfonso Department of Mathematics and Computer Science University of Calabria Ponte Pietro Bucci Rende87036 Italy Department of Environmental Engineering University of Calabria Ponte Pietro Bucci Rende87036 Italy

ISBN: (纸本)9783031812439

We investigate the effectiveness of machine learning (ML) as a complementary tool in enhancing the accuracy of precipitation forecasting in the Calabria region (southern Italy), a Mediterranean area frequently affected by extreme rainfall events, with significant material damage and potential loss of human lives. The operational weather forecasting system developed at the University of Calabria, based on the Weather Research and Forecasting (WRF) model, is considered to this end. We test two ML models, namely an artificial neural network and a random forest, to improve the accuracy of WRF forecasting. We provide several WRF output variables to the ML models. After a proper calibration (learning and test phases) over a two-year-long extended dataset, we assess the results to the original WRF outcomes (validation phase). Preliminary experiments resulted in a significant improvement by the WRF-ML coupled systems. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Impact of data Preprocessing on Wind Forecasting Using machine learning Techniques 4

Impact of Data Preprocessing on Wind Forecasting Using Machi...

引用

4th international conference on Power, Control and Computing Technologies, ICPC2T 2025

作者： Gaikwad, Nidhi Haribhau Mitra, Arghya Payal, Jangilwar Das, Soumyabrata Keshri, Ritesh kumar Department of Electrical Engineering VNIT Nagpur India

ISBN: (纸本)9798331530402

There is a high demand for renewable energy sources as non renewable energy sources are harmful to the environment due to its effects such as global warming. Among non renewable energy source, the wind energy is unpredictable in nature, making Its integration with the grid a concern. By accurately forecasting wind energy, the stability of the grid can be ensured. The aforementioned can be achieved with the help of machine learning (ML) techniques. This paper investigate four ML techniques Multiple Linear Regression, Support Vector Regression, Random Forest Regression, and Decision Trees which are utilized and compared based on the metrics such as root mean square error, mean square error, mean absolute error, and R2 for three different cases. Initially errors are calculated without data preprocessing and later with data preprocessing. It is observed that preprocessing reduces error. This paper also investigates the effect of different seasons on the errors generated by these ML techniques. For this research, data is taken from National Solar Radiance database (NSRD) over two year period from January 2018 to December 2019. © 2025 IEEE.

关键词： Multiple linear regression

来源：评论

学校读者我要写书评

暂无评论

Study on Factors Affecting Carbon Emissions in Ship Navigation and Prediction of EEDI 9th

Study on Factors Affecting Carbon Emissions in Ship Navigati...

引用

9th international conference on Advances in Energy Resources and Environment Engineering, ICAESEE 2023

作者： Zhang, Jingwen Peng, Zitong Liu, Jiaying Yang, Boxin Harbin Engineering University Harbin China University of Southampton Southampton United Kingdom

ISBN: (纸本)9783031665042

In the context of green shipping, this paper analyzes factors influencing carbon emissions during vessel navigation, including wind speed, wind direction, wave conditions, and ocean currents. Through qualitative analysis and data collection from various sources, the study identifies key factors with operational significance that impact vessel navigation speed. A predictive model for shaft power is proposed, outputting Energy Efficiency Existing Ship Index (EEDI) predictions. The research aligns with the trend of low-carbon green shipping, providing insights for reducing operational costs and emission expenses. The study holds significance in the field of artificial intelligence, particularly in machine learning and deep learning. © 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

Slicing Through Bias: Explaining Performance Gaps in Medical Image Analysis Using Slice Discovery Methods 2nd

Slicing Through Bias: Explaining Performance Gaps in Medica...

引用

2nd international Workshop on Fairness of AI in Medical Imaging, FAIMI 2024, and 3rd international Workshop on Ethical and Philosophical Issues in Medical Imaging, EPIMI 2024, Held in Conjunction with the international conference on Medical Image Computing and Computer Assisted Interventions, MICCAI 2024

作者： Olesen, Vincent Weng, Nina Feragen, Aasa Petersen, Eike Technical University of Denmark Kongens Lyngby Denmark Fraunhofer Institute for Digital Medicine MEVIS Bremen Germany

ISBN: (纸本)9783031727863

machine learning models have achieved high overall accuracy in medical image analysis. However, performance disparities on specific patient groups pose challenges to their clinical utility, safety, and fairness. This can affect known patient groups – such as those based on sex, age, or disease subtype – as well as previously unknown and unlabeled groups. Furthermore, the root cause of such observed performance disparities is often challenging to uncover, hindering mitigation efforts. In this paper, to address these issues, we leverage Slice Discovery Methods (SDMs) to identify interpretable underperforming subsets of data and formulate hypotheses regarding the cause of observed performance disparities. We introduce a novel SDM and apply it in a case study on the classification of pneumothorax and atelectasis from chest x-rays. Our study demonstrates the effectiveness of SDMs in hypothesis formulation and yields an explanation of previously observed but unexplained performance disparities between male and female patients in widely used chest X-ray datasets and models. Our findings indicate shortcut learning in both classification tasks, through the presence of chest drains and ECG wires, respectively. Sex-based differences in the prevalence of these shortcut features appear to cause the observed classification performance gap, representing a previously underappreciated interaction between shortcut learning and model fairness analyses. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Contrastive learning

来源：评论

学校读者我要写书评

暂无评论

Smoke Segmentation Improvement Based on Fast Segment Anything Model with YOLOv11 for a Wildfire Monitoring System 10

Smoke Segmentation Improvement Based on Fast Segment Anythin...

引用

10th international conference on Internet of Things, Big data and Security, IoTBDS 2025

作者： Bunpleng, Puchit Thunyatada, Puthtipong Aksornsuwan, Bhutharit Tungpimolrut, Kanokvate Murata, Ken T. Sirindhorn International Institute of Technology Thammasat University Pathum Thani Thailand NECTEC National Science and Technology Development Agency Pathum Thani Thailand National Institute of Information and Communications Technology Tokyo Japan

ISBN: (纸本)9789897587504

Forests and wildlife are crucial parts of our ecosystem. Wildfires occurring in dry and hot regions represent a significant threat to these areas, particularly in ASEAN countries during the dry season. While human observers are often employed to detect wildfires, their scarcity and limited availability highlight the need for automated solutions. This study explores the use of machine learning, specifically computer vision, to enhance wildfire detection by segmenting smoke, an approach which potentially gives information regarding the size and the direction of the spread of the smoke, aiding mitigation efforts. We extend prior work by proposing a model to predict the errors and performance of segmentation masks without access to the ground truth, with the aim of facilitating iterative self-improvement of segmentation models. The FireSpot dataset is used to fine-tune a YOLOv11 model to predict bounding boxes of smoke successfully;subsequently, the outputs of this model are used as a prompt to refine a FastSAM model designed to segment the image into a proposed mask containing the smoke. The proposed mask and the corresponding original image are then used to train a machine learning model where the targets are metrics regarding the error rates of the masks. The results show that a gradient boosting model achieves good prediction performance in predicting some error metrics like the IoU (denoted TPP in this paper) between the proposed and actual segmentation masks with an MSE of 0.03 and R2 of 0.46, as well as the proportion of false positives over the union of the proposed and actual masks (denoted FPP in our paper) with an MSE of 0.0002 and R2 of 0.95, while a pre-trained deep learning model fails to learn the distribution, achieving considerably lower performance for IoU with an MSE of 0.05 and R2 of 0.06 and FPP with an MSE of 0.0002 and R2 of -1.15. These findings open the way to future work where the results of the error prediction model can be used as feedback to

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

FedCSR: A Federated Framework for Multi-Platform Cross-Domain Sequential Recommendation with Dual Contrastive learning 31

FedCSR: A Federated Framework for Multi-Platform Cross-Domai...

引用

31st international conference on Computational Linguistics, COLING 2025

作者： Zheng, Dongyi Zhang, Hongyu Zhai, Jianyang Lin, Zhong Wang, Lingzhi Feng, Jiyuan Liao, Xiangke Tian, Yonghong Xiao, Nong Liao, Qing Pengcheng Laboratory China Sun Yat-sen University China Harbin Institute of Technology Shenzhen China Peking University China

ISBN: (纸本)9798891761964

Cross-domain sequential recommendation (CSR) has garnered significant attention. Current federated frameworks for CSR leverage information across multiple domains but often rely on user alignment, which increases communication costs and privacy risks. In this work, we propose FedCSR, a novel federated cross-domain sequential recommendation framework that eliminates the need for user alignment between platforms. FedCSR fully utilizes cross-domain knowledge to address the key challenges related to data heterogeneity both inter- and intra-platform. To tackle the heterogeneity of data patterns between platforms, we introduce Model Contrastive learning (MCL) to reduce the gap between local and global models. Additionally, we design Sequence Contrastive learning (SCL) to address the heterogeneity of user preferences across different domains within a platform by employing tailored sequence augmentation techniques. Extensive experiments conducted on multiple real-world datasets demonstrate that FedCSR achieves superior performance compared to existing baseline methods. © 2025 Association for Computational Linguistics.

关键词： Contrastive learning

来源：评论

学校读者我要写书评

暂无评论

Stochastic Multiscale Modeling for Thermal Conductivity in Polymeric Graphene-Enhanced Composites: A Study in Interpretable machine learning 30th

Stochastic Multiscale Modeling for Thermal Conductivity in P...

引用

30th international conference on Computational and Experimental Engineering and sciences, ICCES 2024

作者： Liu, Bokai Liu, Pengju Lu, Weizhuo Olofsson, Thomas Intelligent Human-Buildings Interactions Lab Department of Applied Physics and Electronics Umeå University Umeå90187 Sweden Institute of Structural Mechanics Bauhaus-Universität Weimar Marienstr. 15 Weimar99423 Germany

ISBN: (纸本)9783031829062

We've devised an interpretable stochastic integrated machine learning method for predicting thermal conductivity in Polymeric Graphene-Enhanced Composites (PGECs) across different scales. Our approach, integrating techniques like Regression Trees methods (Gradient Boosting machine), handles uncertain parameters via a bottom-up multi-scale framework using Representative Volume Elements in Finite Element Modeling (RVE-FEM). To understand factors affecting predictions, we use the SHapley Additive exPlanations (SHAP) algorithm. This method aligns well with training data, offering promise for efficiently designing new composite materials for thermal management. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Thermal conductivity

来源：评论

学校读者我要写书评

暂无评论

HyperMM: Robust Multimodal learning with Varying-Sized Inputs

HyperMM: Robust Multimodal Learning with Varying-Sized Inpu...

引用

Workshop on Longitudinal Disease Tracking and Modeling with Medical Images and data, LDTM 2024, 5th international Workshop on Multiscale Multimodal Medical Imaging, MMMI 2024, 1st Workshop on machine learning for Multimodal/-sensor Healthcare data, ML4MHD2024 and Workshop on Multimodal learning and Fusion Across Scales for Clinical Decision Support, ML-CDS 2024 held in conjunction with the 27th international conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2024

作者： Chaptoukaev, Hava Marcianó, Vincenzo Galati, Francesco Zuluaga, Maria A. Data Science Department EURECOM Biot France School of Biomedical Engineering and Imaging Sciences Kings College London London United Kingdom

ISBN: (纸本)9783031845246

Combining multiple modalities carrying complementary information through multimodal learning (MML) has shown considerable benefits for diagnosing multiple pathologies. However, the robustness of multimodal models to missing modalities is often overlooked. Most works assume modality completeness in the input data, while in clinical practice, it is common to have incomplete modalities. Existing solutions that address this issue rely on modality imputation strategies before using supervised learning models. These strategies, however, are complex, computationally costly and can strongly impact subsequent prediction models. Hence, they should be used with parsimony in sensitive applications such as healthcare. We propose HyperMM, an end-to-end framework designed for learning with varying-sized inputs. Specifically, we focus on the task of supervised MML with missing imaging modalities without using imputation before training. We introduce a novel strategy for training a universal feature extractor using a conditional hypernetwork, and propose a permutation-invariant neural network that can handle inputs of varying dimensions to process the extracted features, in a two-phase task-agnostic framework. We experimentally demonstrate the advantages of our method in two tasks: Alzheimer’s disease detection and breast cancer classification. We demonstrate that our strategy is robust to high rates of missing data and that its flexibility allows it to handle varying-sized datasets beyond the scenario of missing modalities. We make all our code and experiments available at ***/robustml-eurecom/hyperMM. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Semi-supervised learning

来源：评论

学校读者我要写书评

暂无评论

A Multi-Model Automated Framework for Scalable and Robust Outlier Detection in data Warehousing 5

A Multi-Model Automated Framework for Scalable and Robust Ou...

引用

5th international conference on Advanced Research in Computing, ICARC 2025

作者： Fernando, W.W.A.D.R. Asanka, P.P.G.D. University of Kelaniya Faculty of Science Department of Industrial Management Sri Lanka

ISBN: (纸本)9798331530983

data quality is an important aspect that must be considered to obtain reliable and useful data from data warehouses. This paper presents an automated system using ensemble-based outlier detection methods to detect outliers in large data warehouses. The proposed framework combines several unsupervised machine learning algorithms such as Local Outlier Factor (LOF), Isolation Forest (iForest), One-Class SVM (OCSVM), and Autoencoder to improve the effectiveness of the outlier detection process. Using all these models ensures that the framework solves the drawbacks of single methods in scalability and suitability for large and high-dimensional dynamic data. In addition to functioning as an outlier detection method, the framework employs SHAP (Shapley Additive explanations) to identify the causes of the outliers to provide a more in-depth analysis of such outliers. This capability enables organizations to see patterns, possible areas of inefficiency that lead to an outlier, and areas of data inconsistency that need to be corrected. The effectiveness of the system has been verified using the AdventureWorksDW2019 data warehouse, showcasing its resilience in detecting intricate irregularities and tracking their sources across many fact tables. This study provides useful recommendations for enhancing data quality in a range of data warehousing applications and lays the groundwork for scalable outlier identification. © 2025 IEEE.

关键词： Unsupervised 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：