检索结果-内蒙古大学图书馆

Beyond administrative reports: a deep learning framework for classifying and monitoring crime and accidents leveraging large-scale online news

引用

Neural Computing and Applications 2025年第10期37卷 7183-7205页

作者： Tuarob, suppawong Tatiyamaneekul, Phonarnun Pongpaichet, siripen Tawichsri, Tanisa Noraset, Thanapon Faculty of Information and Communication Technology Mahidol University Nakhon Pathom Thailand Department of Computer Science University College London London United Kingdom Puey Ungphakorn Institute for Economic Research Bank of Thailand Bangkok Thailand

The escalating prevalence of violent crimes and accidents underscores the urgent need for efficient and timely monitoring systems. Traditional methods reliant on administrative reports often suffer from significant delays. This paper proposes CRIMsON, a novel framework that leverages large-scale online news to provide real-time insights into crime and accident trends. CRIMsON utilizes a multi-label classification technique that leverages a fine-tuned, pre-trained, cross-lingual language model to accurately categorize news articles. Our experimental results, conducted on a substantial dataset of Thai news articles, demonstrate superior performance, achieving an average F1 score of 86%. Beyond classification, CRIMsON aggregates categorized news into real-time statistics, revealing strong correlations between news-reported incidents and official crime data. This study pioneers online news as a reliable and timely crime and accident monitoring source, offering valuable insights for law enforcement, policymakers, and researchers. © The Author(s) 2025.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Prioritization Hindsight Experience based on spatial Position Attention for Robots

引用

Machine Intelligence research 2025年第1期22卷 160-175页

作者： Ye Yuan Yu sha Feixiang sun Haofan Lu shuiping Gou Jie Luo China Nuclear Power Technology Research Institute Co. Ltd.Shenzhen518028China School of Artificial Intelligence Xidian UniversityXi’an710071China Suzhou Tianhe China Power International Engineering Technology Co. Ltd.Suzhou215000China

sparse rewards pose significant challenges in deep reinforcement learning as agents struggle to learn from experiences with limited reward *** experience replay(HER)addresses this problem by creating“small goals”within a hierarchical decision ***,HER does not consider the value of different episodes for agent *** this paper,we propose sPAHER,a framework for prioritizing hindsight experiences based on spatial position *** allows the agent to prioritize more valuable experiences in a manipulation *** achieves this by calculating transition and trajectory spatial position functions to determine the value of each episode for experience *** evaluate sPAHER on eight robot manipulation tasks in the Fetch and Hand environments provided by OpenAI *** results show that our method improves the final mean success rate by an average of 3.63%compared to HER,especially in challenging Hand ***,these improvements are achieved without any increase in computation time.

关键词： Hindsight experience replay spatial position attention sparse reward deep reinforcement learning prioritization hindsight experience

来源：评论

学校读者我要写书评

暂无评论

A Real-World segmentation Model for Melanocytic and Nonmelanocytic Dermoscopic Images

A Real-World Segmentation Model for Melanocytic and Nonmelan...

引用

18th International Joint Conference on Biomedical Engineering systems and Technologies, BIOsTEC 2025

作者： Melissa, Eleonora Riabitch, Daria Lazzeri, Linda La Rosa, Federica Benvenuti, Chiara D’acunto, Mario Bagnoni, Giovanni Massi, Daniela Laurino, Marco Institute of Clinical Physiology National Research Council Pisa Italy Uniti of Dermatologia Specialist Surgery Area Department of General Surgery Livorno Hospital Azienda Usl Toscana Nord Ovest Italy Institute of Biophysics National Research Council Pisa Italy Department of Health Sciences Section of Pathological Anatomy University of Florence Florence Italy

segmentation is a critical step in computer-aided diagnosis (CAD) systems for skin lesion classification. In this study, we applied the Deeplabv3+ network to segment real dermoscopic images. The model was trained on public datasets and tested both on public and on a disjoint set of images from the TELEMO project, covering six clinically significant skin lesion types: basal cell carcinoma, squamous cell carcinoma, melanoma, benign nevi, actinic keratosis and seborrheic keratosis. Our model achieved a testing global accuracy of 97.88% on public dataset and of 92.62% on TELEMOdataset, outperformingliterature models. Although some misclassi f ications occurred, largely due to class imbalance, the model demonstrated strong generalization to real-world clinical images, a critical achievement for deep learning in medical imaging. To evaluate the clinical relevance of our segmentation, we extracted ten key features related to lesion border and diameter. Notably, the "Diame ters Mean" and "Area to Perimeter Product" features revealed significant differences between melanoma-nevi and basal cell carcinoma-nevi classes, with strong effect sizes. These findings suggest that border features are crucial for distinguishing between multiple skin lesion types, highlighting the model’s potential for aiding dermatological diagnoses. © 2025 by sCITEPREss– science and Technology Publications, Lda.

关键词： Medical imaging

来源：评论

学校读者我要写书评

暂无评论

A Cost-Aware Adaptive Bike Repositioning Agent Using Deep Reinforcement learning

引用

IEEE TRANsACTIONs ON INTELLIGENT TRANsPORTATION sYsTEMs 2025年第4期26卷 4923-4933页

作者： staffolani, Alessandro Darvariu, Victor-Alexandru Bellavista, Paolo Musolesi, Mirco Univ Bologna Dept Comp Sci & Engn I-40126 Bologna Italy ISTI CNR Natl Council Res Italy I-56127 Pisa Italy Univ Oxford Oxford Robot Inst Dept Engn Sci Oxford OX1 3PJ England UCL Dept Comp Sci London WC1E 6BT England

Bike sharing systems (Bss) represent a sustainable and efficient urban transportation solution. A major challenge in Bss is repositioning bikes to avoid shortage events when users encounter empty or full bike lockers. Existing algorithms unrealistically rely on precise demand forecasts and tend to overlook substantial operational costs associated with reallocations. This paper introduces a novel Cost-aware Adaptive Bike Repositioning Agent (CABRA), which harnesses advanced deep reinforcement learning techniques in dock-based Bss. By analyzing demand patterns, CABRA learns adaptive repositioning strategies aimed at reducing shortages and enhancing truck route planning efficiency, significantly lowering operational costs. We perform an extensive experimental evaluation of CABRA utilizing real-world data from Dublin, London, Paris, and New York. The reported results show that CABRA achieves operational efficiency that outperforms or matches very challenging baselines, obtaining a significant cost reduction. Its performance on the largest city comprising 1765 docking stations highlights the efficiency and scalability of the proposed solution even when applied to Bss with a great number of docking stations.

关键词： Costs Vehicle dynamics Urban areas Docking stations Deep reinforcement learning scalability Optimization Electronic mail computer science shared transport Dynamic bike repositioning reinforcement learning resource allocation

来源：评论

学校读者我要写书评

暂无评论

Deep learning enabled computer vision in remanufacturing and refurbishment applications: defect detection and grading for smart phones

引用

Journal of Remanufacturing 2025年第1期15卷 65-95页

作者： Mohandas, Reenu southern, Mark Fitzpatrick, Colin Hayes, Martin Department of Electronic and Computer Engineering University of Limerick Limerick Limerick V94 T9PX Ireland School of Engineering and Enterprise Research Centre University of Limerick Limerick Limerick V94 T9PX Ireland

This work demonstrates the use of Deep learning-based computer Vision for Remanufacturing end-of-life consumer electronics products, considering smartphones as the use-case. We implemented automated detection of screen defects such as scratches and cracks. In turn, this could lead to increased reuse of smartphones in a secondary market alongside new ones to meet consumer demand. The refurbishment of smartphone devices is a growing industry heavily dependent on manual labor, making decisions subjective, especially in grading the severity of damage. A critical analysis of defect detection and smartphone grading from a remanufacturing perspective is conducted. This analysis is significant in a remanufacturing sector that deals with dynamically varying input of cores (used products for remanufacturing), characteristics, timing, and number of returns. The solution we propose here is novel in its own right, requiring data analysis and collection, data curing, defect parameterization, and dataset building to enable model-based training and detection experiments. We collected and annotated a dataset to detect and grade the various defects based on their severity. A range of deep learning models was trained on the dataset to obtain baseline results for the state-of-the-art deep learning detection models, including YOLOv7, YOLOv8, YOLO11 variants, and Mask R-CNN. Our experiments also showed improved precision values when the pre-trained models were pre-fine-tuned using a road crack segmentation dataset before training on our phone defect dataset. The inference time for the YOLOv8x model is 8ms. This reduced inference time with a high precision of 70.4% indicates that a consistent, fast, and accurate grading is achieved here, ensuring a high throughput rate in the remanufacturing process and ensuring sustainability. © The Author(s) 2025.

关键词： computer vision Deep learning Defect detection Refurbishment smartphone remanufacturing YOLO

来源：评论

学校读者我要写书评

暂无评论

A deep convolutional neural network-based keratitis detection and classification approach

引用

Multimedia Tools and Applications 2025年 1-21页

作者： susila, s. Janet Grace Kavitha, D. Department of Computer Science St. Peter’S Institute of Higher Education and Research Chennai India Department of Computer Applications St. Peter’S Institute of Higher Education and Research Chennai India

Keratitis is an intraocular disease that is an inflammation of the cornea and one of the challenges of ophthalmology due to its multifarious character and mild initial symptoms. In this paper, we present KeraNet Deep Convolutional Neural Network (DCNN), which is a novel deep learning architecture to aid the early detection of keratitis. Our proposed model adds on a hybrid attention mechanism that employs spatial and channel attention to guide the model’s focus to the relevant parts of the cornea, along with adaptive multi-scale feature extraction, which helps in capturing the features of lesions of different sizes. The architecture uses Hierarchical Dense Residual Blocks (HDRBs) to improve feature extraction without the problem of vanishing gradients. Deformable Convolutions, on the other hand, have the advantage of changing receptive fields in order to successfully configure the abnormal formations. In our experiments, KeraNet DCNN showed a high balanced accuracy of 98.4% and a sensitivity of 98.7%, which is better performance than all existing approaches, including EfficientNetV2-M and OL-CDACNN, in terms of sensibility and specificity. The image analysis tools capabilities were further tested on a new model containing a loss that combines the localization of lesions and the prediction, hence optimally addressing the problem of prediction accuracy and interpretation. Having undergone extensive testing and evaluation, KeraNet DCNN shows good promise in the detection of keratitis at its earliest stage;this is a revolutionary breakthrough in the field of automated diagnosis in ophthalmology. © The Author(s), under exclusive licence to springer science+Business Media, LLC, part of springer Nature 2025.

关键词： Ophthalmology

来源：评论

学校读者我要写书评

暂无评论

A merged machine learning model for seasonal climate prediction in China

引用

science China Earth sciences 2025年第5期68卷 1469-1484页

作者： Danwei QIAN Yanyan HUANG Huijun WANG State Key Laboratory of Climate System Prediction and Risk Management/Key Laboratory of Meteorological Disaster Ministry of Education/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters Nanjing University of Information Science and Technology Southern Marine Science and Engineering Guangdong Laboratory (Zhuhai) Nansen-Zhu International Research Centre Institute of Atmospheric Physics Chinese Academy of Sciences

Improving the current level of skill in seasonal climate prediction is urgent for achieving sustainable socioeconomic development, and this is especially true in China where meteorological disasters are experienced frequently. In this study, based upon big climate data and traditional statistical prediction experiences, a merged machine learning model(Y-model) was developed to address this, as well as to further explore unknown potential predictors. In Y-model, empirical orthogonal function analysis was firstly applied to reduce the data dimensionality of the target predictand(temperature and precipitation in the four seasons over China). Image recognition techniques were used to automatically identify possible predictors from the big climate data. These predictors, associated with significant circulation anomalies, were recombined into a large ensemble according to different threshold settings for five factors determining the statistical forecast skill. Facebook Prophet was chosen to conduct the independent hindcasts for each season's climate at a lead time of two months. During 2011–2022, the seasonal climate in China was skillfully predicted by Y-model, with an averaged pattern correlation coefficient skill of 0.60 for temperature and 0.24 for precipitation, outperforming CFsv2. Potential predictor analysis for recent extreme events suggested that prior signals from the Indian Ocean and the stratosphere were important for determining the super Mei-yu in 2020, while the prior sea surface temperature over the western Pacific and the soil temperature over West Asia may have contributed to the extreme high temperatures in 2022. Our study provides new insights for seasonal climate prediction in China.

关键词： Machine learning seasonal climate prediction China’s climate Image recognition techniques Large ensemble

来源：评论

学校读者我要写书评

暂无评论

A hybrid deep learning model for UWB radar-based human activity recognition

引用

INTERNET OF THINGs 2025年 29卷

作者： Khan, Irfanullah Guerrieri, Antonio serra, Edoardo spezzano, Giandomenico Univ Calabria Via P Bucci I-87036 Arcavacata Di Rende CS Italy Natl Res Council Italy ICAR CNR Inst High Performance Comp & Networking Via P BucciCubo 8-9C I-87036 Arcavacata Di Rende CS Italy Boise State Univ 1910 W Univ Dr Boise ID 83725 USA

In today's world, energy efficiency in buildings has become a top priority due to the significant energy waste caused by the operation of inefficient electrical appliances. Conventional methods of reducing energy waste cause discomfort for occupants inside buildings. One promising way to optimize energy consumption is to synchronize appliance operation with building occupants' dynamic behavior. Internet of Things (IoT) technologies, which allow for widespread data collection and execution of Machine learning (ML) algorithms, enabled the creation of smart Buildings (sBs). sBs can learn patterns from the inhabitant's behavior residing in, and adjust their operations in accordance with these behaviors. By doing so, these sBs could reduce energy waste, enhancing resource efficiency and consequently reduce CO2 gas emissions. Furthermore, they could improve the overall comfort of the living environment and help with sustainability initiatives. In this context, this paper proposes a novel approach that uses a hybrid deep-learning model to recognize complex human activities based on data collected from ultra-wideband (UWB) radar technology. Our approach, called Hybrid Deep learning Model for Activity Recognition (HDL4AR), includes long-short-term memory (LsTM) and a one-dimensional convolutional neural network (1D-CNN). We deploy a real-time case study by collecting data from 22 participants involved in 10 diverse activities at the headquarters of the ICAR-CNR in the IoT Laboratory, Italy. Moreover, we conducted a comprehensive benchmark of the HDL4AR approach against various statistical techniques and other deep learning models recently introduced in the literature. Results show that our proposed approach outperformed conventional methods and achieved an impressive accuracy of 98.42%.

关键词： Internet of Things smart buildings Human activity recognition UWB radar Artificial intelligence Neural networks LsTM

来源：评论

学校读者我要写书评

暂无评论

A Collaborative Change Detection Network based on Feature Pyramids 5

A Collaborative Change Detection Network Based on Feature Py...

引用

5th IEEE International Conference on Power, Electronics and computer Applications, ICPECA 2025

作者： Wu, Gang Liu, Luyao Liu, Yigao Zhang, Zenghui Cong, Li Huang, Chengbin Lv, Ning State Grid Jilin Province Electric Power Company Changchun China Engineering Technology Research Institute Xidian University Xi'an China

ISBN: (纸本)9798331533694

The application of high-resolution satellite images in deep learning-based change detection methods has become increasingly popular. However, the down-sampling and cropping strategies deployed to fit the GPU memory constraint often result in incomplete semantic information and loss of fine details. This paper proposes a collaborative network based on feature pyramids, which enables networks to learn local and global features from cropped and down-sampled image patches. A feature-sharing mechanism is also introduced to fuse global features and local features. We evaluated the network on the LEVIR and s2Looking datasets by comparing it with some representative change detection networks. Comparative experiments show that the proposed network performs better in multi-scale change detection, improving precision by 2.69% on the LEVIR-CD dataset, and improving precision and recall by 6.83% and 2.68% on the s2Looking dataset, respectively. © 2025 IEEE.

关键词： Change detection

来源：评论

学校读者我要写书评

暂无评论

Assessing the impact of differential privacy in transfer learning with deep neural networks and transformer language models

引用

Neural Computing and Applications 2025年第6期37卷 5097-5119页

作者： sousa, samuel Trügler, Andreas Kern, Roman Know Center Research GmbH Graz Austria Graz University of Technology Graz Austria Graz Center for Machine Learning Graz Austria University of Graz Graz Austria

The realization of trustworthy artificial intelligence strongly relies on privacy, fairness, and accountability requirements. Although model trustworthiness results from the synergy between these requirements, some effects often clash with one another and may propagate issues into downstream tasks. This paper empirically indicates that optimization algorithms with differential privacy (DP), such as differentially private stochastic gradient descent (DP-sGD) and differentially private Adam (DP-Adam), disproportionately increase differences in accuracy between groups and domains in transfer learning based on convolutional neural network (CNN) and long-short-term memory (LsTM) models. We investigate the impact of DP-sGD and DP-Adam in transfer learning models, which receive DP noise to hinder privacy risks and perform predictions on target domain data. In addition, after extensive experimental evaluation, we provide evidence that fine-tuning may reduce the differences in accuracy for most groups compared to gradient boosting and ensemble methods. However, differentially private optimizers may reverse gains in accuracy from transfer learning for CNNs and LsTMs. Unevenly reduced model accuracy may jeopardize the predictions’ fairness, causing numerous real-world implications, such as denying credit to people in need or undermining complex language processing. Identifying scenarios where privacy protection conflicts with fairness for transfer learning is a problem often overlooked yet directly aligned with the accountability requirement in trustworthy artificial intelligence. Therefore, assessing the impact of differentially private optimizers in transfer learning is the first step to developing mitigation strategies. © The Author(s), under exclusive licence to springer-Verlag London Ltd., part of springer Nature 2024.

关键词： Differential privacy

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：