检索结果-内蒙古大学图书馆

Interpretable vertical federated learning with privacy-preserving multi-source data integration for prognostic prediction

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2025年 148卷

作者： Wang, Qingyong Anhui Agr Univ Sch Informat & Artificial Intelligence Hefei Peoples R China

Digital healthcare is entering a new era characterized by the availability of extensive biomedical data. This wealth of data not only facilitates clinical diagnosis research but also enables the customization of appropriate treatments by incorporating various aspects of patient data. Advanced deep learning methods have the potential to improve clinical diagnosis and treatment efficient significantly. However, a major challenge lies in collecting datasets that are sufficiently large and diverse to train robust prognostic prediction models. Although various types of multi-source data can provide comprehensive insights for medical diagnosis, patient data are often collected by different institutions. Collaborations among multiple institutions involve sharing such data, which raises significant concerns regarding privacy protection and ownership. To address these challenges, this study proposes an interpretable vertical federated learning with meta-optimization (VerFedMeo) that integrates private multi-source data and improves the accuracy of cancer prognosis prediction. Specifically, VerFedMeo employs a partial likelihood function approximation as a its learning objective and derives gradient formulas to achieve more accurate approximations. Meta-optimization has been developed to enhance the model's learning capacities by automatically identifying optimal hyperparameters. The experimental results demonstrate that VerFedMeo approach significantly outperforms existing methods, achieving improvements of 5.8% in F1-score and 3.5% in area under the curve(AUC) compared to the suboptimal method. Moreover, we presented an interpretable consequence of the potential summaries that could be used by clinicians to predict disease prognosis.

关键词： Deep learning Vertical federated learning Meta-optimization multi-source data integration Interpretable disease prognosis

来源：评论

学校读者我要写书评

暂无评论

Innovation and Growth of multi-source data integration in Visual Communication Design

Innovation and Growth of Multi-Source data Integration in Vi...

引用

2023 International Conference on Evolutionary Algorithms and Soft Computing Techniques, EASCT 2023

作者： Zhang, Yingyu Sun, Hongbo Changchun Humanities College Visual Communication Design Changchun China

ISBN: (纸本)9798350313413

In visual communication design (VCD), multi-source data integration plays an important role in innovation and growth. This paper studies and discusses the means of information dissemination in the new media era by analyzing typical digital interaction methods and combining current social hotspots. With the guiding ideology of 'multiple sources', this paper proposes a brand-new, efficient method that meets the needs of the public, aiming to achieve a visual communication art form with multiple expressive effects and a design language system with rich connotations. In the process of conception, this paper covers the technical level of image digital conversion and fusion and interactive data integration, designs a VCD model, and conducts testing and research on multi-source data integration algorithms. The test results show that the data processing time of the multi-source data integration algorithm is within 6 seconds, and the fastest time is 3 seconds. The application of these technologies makes it possible to continuously change and enrich the content and form of VCD. The research on the integration of multi-source data in visual communication design covers multiple important findings. The research needs to explore data processing and integration methods in order to effectively process and integrate data from different data sources. It also considers users' feedback and evaluation of multi-source data integration visual communication design to understand their perception and attitude towards design effectiveness, usability, and information presentation. © 2023 IEEE.

关键词： multi-source data integration visual communication visual design visual innovation

来源：评论

学校读者我要写书评

暂无评论

Design and implementation of a fast integration method for multi-source data in high-speed network

引用

JOURNAL OF HIGH SPEED NETWORKS 2023年第3期29卷 251-263页

作者： Ma, Lei Zhang, Yanning Garcia-Diaz, Vicente Beijing Polytech Sch Telecommun Engn Beijing Peoples R China Univ Oviedo Dept Comp Sci Oviedo Spain

The data collected by the distributed high-speed network has multiple sources. Therefore, in order to realize the rapid integration of multi-source data, this paper designs a rapid data integration method based on the characteristics of the distributed high-speed network. First, we use linear regression analysis to build a distributed perceptual data model, so that network nodes can only transmit the parameter information of the regression model, so as to simplify the data collection. Then, a dead band amplitude limiting nonlinear link is added at the high frequency channel side to filter and assimilate the data. Finally, the data feature vectors are extracted as the training samples of the neural network to obtain the mapping relationship between different feature vectors, and then the decision level data integration is achieved by training the neural network. The experimental results show that this method can accurately collect high-speed network data, and the data collection deviation is always less than 5 mu rad;This method has good filtering effect on data and can eliminate the interference of burr signal;The convergence speed of this method is fast, and the data assimilation can be completed within 0.4 s, which is conducive to improving the speed of data integration;With the increase of network size, the average traffic load of this method increases less.

关键词： Distributed system data acquisition high-speed network multi-source data integration nonlinear filtering heterogeneous assimilation

来源：评论

学校读者我要写书评

暂无评论

data-driven safety management of worker-equipment interactions using visual relationship detection and semantic analysis

引用

AUTOMATION IN CONSTRUCTION 2025年 175卷

作者： Liu, Yipeng Wang, Junwu Torbaghan, Mehran Eskandari Wuhan Univ Technol Sanya Sci & Educ Innovat Pk Sanya 572025 Peoples R China Wuhan Univ Technol Sch Civil Engn & Architecture Wuhan 430070 Peoples R China Univ Birmingham Dept Civil Engn Birmingham B15 2TT England

Existing technologies struggle to accurately identify interactions between workers and equipment, as well as the deep semantics of complex construction scenes. To address these limitations, this paper proposes an automated construction site safety management system designed to enhance scene understanding and identify safety hazards while focusing on hazard-area and personal protective equipment (PPE) interaction. The system transforms image information into worker-centric triplets and generates precise textual descriptions through semantic enhancement, enabling effective scene analysis. By comparing the generated descriptions with predefined hazard statements, the system identifies potential risks. Experimental results demonstrate a 9.6 % improvement in recall for Ng-mR@K metrics (K = 20, 50, 100). Additionally, the system successfully filters over 90 % of invalid relationships, achieving 83.7 % accuracy in semantic similarity matching, significantly enhancing detection precision and semantic understanding. By advancing from object detection to a structured image-to-triplet-to-text framework, this paper offers an efficient and reliable solution for automated construction site safety management.

关键词： Construction site safety multi-source data integration Visual relationship detection Semantic triplets Image-to-text transformation Semantic enhancement

来源：评论

学校读者我要写书评

暂无评论

Path-ComDim multi-block data analysis in the cachaça investigation: Integrating sensory analysis with electronic and vibrational spectroscopies

引用

FOOD CHEMISTRY 2025年 481卷 143894页

作者： Cruz, Eduarda Barbieri da Silva Baqueta, Michel Rocha Rutledge, Douglas N. Valderrama, Patricia Univ Tecnol Fed Parana Campus Campo Mourao UTFPR CM Campo Mourao Parana Brazil Univ Paris Saclay Orsay France CNRS Museum Natl Hist Nat MCAM UMR7245 Paris France

Cacha & ccedil;a consumption is valued for its unique characteristics, many of which stem from aging in various wooden barrels, imparting specific flavors and aromas. However, whether unaged cacha & ccedil;as with artificial flavoring exhibit attributes similar to those of naturally aged cacha & ccedil;as remains unclear. This study examined sensory and chemical aspects of both aged and flavored cacha & ccedil;as through an integrated approach combining sensory analysis with ultraviolet, visible, mid-infrared, and near-infrared spectroscopies. In the chemometric domain, PathComDim was employed to integrate and interpret multi-source data, revealing sensory-spectral correlations to characterize cacha & ccedil;a samples derived from aging and flavoring processes. Although the production regions of the cacha & ccedil;as were known beforehand, this factor did not distinguish them. Overall, the findings indicate that flavored and aged cacha & ccedil;as may share sensory similarities because some flavored cacha & ccedil;as can exhibit sensory attributes similar to those aged in Oak, Balsamo, and Umburana woods.

关键词： Brazilian cacha & ccedil a Chemometrics multi-source data integration Sensory dimensions Spectral techniques Sugarcane spirits

来源：评论

学校读者我要写书评

暂无评论

Digital twin comprehensive models: a study of ancient tree ecological environment quality assessment based on a cyber-physical system

引用

ENVIRONMENTAL MONITORING AND ASSESSMENT 2025年第4期197卷 1-28页

作者： Chen, Yansheng Huang, Huazhi Li, Jie Zheng, Zejiong Gao, Fengjun Han, Xiaoge Gao, Yanglin Guangdong Ind Polytech Univ Dept Inst Big Data & Artificial Intelligence Appli Guangzhou 510300 Peoples R China Guangdong Ind Polytech Sch Ecoenvironm Technol Guangzhou 510300 Peoples R China Guangdong Ind Polytech Sch Finance & Trade Guangzhou 510300 Peoples R China

This study leverages cyber-physical system (CPS) technology to create a digital twin model for assessing the ecological quality of ancient trees. Integrating multi-source data and machine learning, our model provides tailored conservation strategies, supports ecological restoration, and enhances disaster response capabilities. Key findings illustrate that the model is precise in monitoring tree health, managing water resources, and predicting the impacts of natural disasters. This innovative approach provides significant advantages in real-time monitoring and long-term ecological management, ensuring the sustainability of ancient tree ecosystems. Our results highlight the model's potential to transform ecological conservation practices and offer a reliable tool for researchers and practitioners in environmental science.

关键词： Ancient tree conservation Real-time monitoring Machine learning in ecology Ecosystem sustainability Habitat quality assessment multi-source data integration

来源：评论

学校读者我要写书评

暂无评论

Optimization of Dynamic Pricing Models for Consumer Segmentation Markets and Analysis of Big data-Driven Marketing Strategies

引用

JOURNAL OF ORGANIZATIONAL AND END USER COMPUTING 2025年第1期37卷

作者： Zhang, Qi Shi, Qiang Alatas, Bilal Yuan, Yu-His ANHUI Int Studies Univ Hefei Peoples R China Firat Univ Elazig Turkiye Natl Taiwan Univ Arts Taipei Taiwan

In response to the challenges posed by globalization and rapid technological advancements, traditional static pricing models are no longer sufficient to capture the dynamic nature of consumer behavior and market fluctuations. This study proposes a "multi-dimensional Dynamic Pricing Optimization and Consumer Behavior Prediction Model Driven by Big data," which integrates multi-source data and reinforcement learning to improve dynamic pricing strategies. Through a hybrid model architecture using Random Forest and LSTM, it captures both static and time-series features. Experimental results show that the proposed model significantly outperforms baseline models, achieving a 43% reduction in Mean Squared Error (MSE), a 28% decrease in Mean Absolute Percentage Error (MAPE), a 6.5% increase in Accuracy, and a 14.7% increase in Cumulative Revenue. These findings confirm the model's ability to enhance prediction accuracy, optimize pricing strategies, and maximize revenue, demonstrating its potential for real-world applications in industries like e-commerce, finance, and advertising.

关键词： Dynamic Pricing Consumer Behavior Prediction Reinforcement Learning multi-source data integration Real-Time Strategy Optimization

来源：评论

学校读者我要写书评

暂无评论

Enhancing keratoconus detection with transformer technology and multi-source integration

引用

ARTIFICIAL INTELLIGENCE REVIEW 2024年第1期58卷 1-31页

作者： Ismael, Osama Cairo Univ Fac Comp & Artificial Intelligence Cairo Egypt

Keratoconus is a progressive eye disease characterized by the thinning and conical distortion of the cornea, leading to visual impairment. Early and accurate detection is essential for effective management and treatment. Traditional diagnostic methods, relying primarily on corneal topography, often fail to detect early-stage keratoconus due to their subjective nature and limited scope. In this research, we present a novel multi-source detection approach utilizing transformer technology to predict keratoconus progression more accurately. By integrating and analyzing diverse data sources, including corneal topography, aberrometry, pachymetry, and biomechanical properties, our method captures subtle changes indicative of disease progression. Transformer networks, known for their capability to model complex dependencies in data, are employed to handle the multimodal datasets effectively. Experimental results demonstrate that our approach significantly outperforms existing methods, such as SVM-based, Random Forests-based, and CNN-based models, in terms of accuracy, precision, recall, and F-score. Moreover, the proposed system exhibits lower execution times, highlighting its efficiency in clinical settings. This innovative methodology holds the potential to revolutionize keratoconus management by enabling earlier and more precise interventions, ultimately enhancing patient outcomes and contributing significantly to both the medical and machine learning communities.

关键词： Keratoconus progression Transformers technology multi-source data integration Deep learning Corneal topography

来源：评论

学校读者我要写书评

暂无评论

Study on the Influencing Factors of Forest Tree-Species Classification Based on Landsat and Sentinel-2 Imagery

引用

FORESTS 2024年第9期15卷 1511页

作者： Lai, Xin Tang, Xu Ren, Zhaotong Li, Yuecan Huang, Runlian Chen, Jianjun You, Haotian Guilin Univ Technol Coll Geomatics & Geoinformat 12 Jiangan Rd Guilin 541006 Peoples R China Guangxi Forest Inventory & Planning Inst 14 Zhonghua Rd Nanning 530011 Peoples R China

Accurate forest tree-species classification not only provides data support for forest resource management but also serves as a crucial parameter for simulating various ecological processes. However, the results of forest tree-species classification have been affected by multiple factors, such as the spectral resolution, spatial resolution, and radiometric resolution of imagery, the classification algorithms used, the sample size, and the timing of image acquisition phases. Although there are many studies on the impact of individual factors on tree-species classification, there is a lack of systematic studies quantifying the magnitude of these factors' influences, leading to uncertainties about the relative importance of different factors. In this study, Landsat-8, Landsat-9, and Sentinel-2 imagery was used as the foundational data, and random forest (RF), gradient tree boosting (GTB), and support vector machine (SVM) algorithms were employed to classify forest tree species. High-accuracy regional forest tree-species classification was achieved by exploring the impacts of spectral resolution, spatial resolution, radiometric resolution, classification algorithms, sample size, and image time phases. The results show that, for the commonly used Landsat-8, Landsat-9, and Sentinel-2 imagery, the tree-species classification results from Landsat-9 are the best, with an overall accuracy of 74.21% and a kappa of 0.71. Among the various influencing factors, the classification algorithm, image time phases, and sample size have relatively larger impacts on tree-species classification results, each exceeding 10%, while the positive impact of radiometric resolution is the smallest, at only 3.15%. Conversely, spectral and spatial resolutions had negative effects on tree-species classification results, at -4.09% and -1.4%, respectively. Based on the 30-m spring Landsat-9 and Sentinel-2 imagery, with 300 samples for each tree-species category, the classification results using the RF

关键词： Landsat-8 Landsat-9 Sentinel-2 multi-source data integration tree-species classification influencing factors

来源：评论

学校读者我要写书评

暂无评论

A graph-based neural network approach to integrate multi-source data for urban building function classification

引用

COMPUTERS ENVIRONMENT AND URBAN SYSTEMS 2024年 110卷

作者： Kong, Bo Ai, Tinghua Zou, Xinyan Yan, Xiongfeng Yang, Min Wuhan Univ Sch Resource & Environm Sci Wuhan 430079 Peoples R China Tongji Univ Coll Surveying & Geoinformat Shanghai 200092 Peoples R China

Accurately understanding the functions of buildings is crucial for urban monitoring, analysis of urban economic structures, and effectively allocating resources. Previous studies have investigated building function classification using single or dual data sources. However, the complexity of building functions cannot be fully reflected by a limited number of data sources. In addition, the functions of adjacent buildings often exhibit correlations, and contextual information between buildings has been ignored in previous studies. To address these problems, we propose a graph-based neural network (GNN) approach for building function classification that integrates multisource data and mines contextual information between buildings. This approach initially extracts four types of features related to building functions: morphological features from vector-buildings, visual features from streetview images, spectral features from satellite images, and socio-economic features from points of interest. The buildings are then modeled as a graph, where the nodes and edges represent the buildings and their proximity. Descriptive features of the nodes were obtained by concatenating the aforementioned features. Finally, the constructed graph was fed into the GraphSAmple and aggreGatE (GraphSAGE) model, which is a typical GNN model for building function classification. The experimental results showed that our approach achieved an F1score of 91.0%, which was 10.3-12.6% higher than that of the three comparison approaches. In addition, ablation experiments using different data sources revealed that the four data sources were complementary to each other and contributed to improving the building function classification. Our strategy provides an alternative and efficient solution for building function classification.

关键词： Building function Classification multi-source data integration Graph-based neural network

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：