Speech-face association aims to achieve identity matching between facial images and voice segments by aligning cross-modal features.Existing research primarily focuses on learning shared-space representations and comp...Speech-face association aims to achieve identity matching between facial images and voice segments by aligning cross-modal features.Existing research primarily focuses on learning shared-space representations and computing one-to-one similarities between cross-modal sample pairs to establish their correlation.However,these approaches do not fully account for intra-class variations between the modalities or the many-to-many relationships among cross-modal samples,which are crucial for robust association modeling.To address these challenges,we propose a novel framework that leverages global information to align voice and face embeddings while effectively correlating identity information embedded in both modalities.First,we jointly pre-train face recognition and speaker recognition networks to encode discriminative features from facial images and voice segments.This shared pre-training step ensures the extraction of complementary identity information across modalities.Subsequently,we introduce a cross-modal simplex center loss,which aligns samples with identity centers located at the vertices of a regular simplex inscribed on a hypersphere.This design enforces an equidistant and balanced distribution of identity embeddings,reducing intra-class variations.Furthermore,we employ an improved triplet center loss that emphasizes hard sample mining and optimizes inter-class separability,enhancing the model’s ability to generalize across challenging scenarios.Extensive experiments validate the effectiveness of our framework,demonstrating superior performance across various speech-face association tasks,including matching,verification,and retrieval.Notably,in the challenging gender-constrained matching task,our method achieves a remarkable accuracy of 79.22%,significantly outperforming existing approaches.These results highlight the potential of the proposed framework to advance the state of the art in cross-modal identity association.展开更多
Emotion recognition under uncontrolled and noisy environments presents persistent challenges in the design of emotionally responsive systems.The current study introduces an audio-visual recognition framework designed ...Emotion recognition under uncontrolled and noisy environments presents persistent challenges in the design of emotionally responsive systems.The current study introduces an audio-visual recognition framework designed to address performance degradation caused by environmental interference,such as background noise,overlapping speech,and visual obstructions.The proposed framework employs a structured fusion approach,combining early-stage feature-level integration with decision-level coordination guided by temporal attention mechanisms.Audio data are transformed into mel-spectrogram representations,and visual data are represented as raw frame sequences.Spatial and temporal features are extracted through convolutional and transformer-based encoders,allowing the framework to capture complementary and hierarchical information fromboth sources.Across-modal attentionmodule enables selective emphasis on relevant signals while suppressing modality-specific noise.Performance is validated on a modified version of the AFEW dataset,in which controlled noise is introduced to emulate realistic conditions.The framework achieves higher classification accuracy than comparative baselines,confirming increased robustness under conditions of cross-modal disruption.This result demonstrates the suitability of the proposed method for deployment in practical emotion-aware technologies operating outside controlled environments.The study also contributes a systematic approach to fusion design and supports further exploration in the direction of resilientmultimodal emotion analysis frameworks.The source code is publicly available at https://github.com/asmoon002/AVER(accessed on 18 August 2025).展开更多
Detecting surface defects on unused rails is crucial for evaluating rail quality and durability to ensure the safety of rail transportation.However,existing detection methods often struggle with challenges such as com...Detecting surface defects on unused rails is crucial for evaluating rail quality and durability to ensure the safety of rail transportation.However,existing detection methods often struggle with challenges such as complex defect morphology,texture similarity,and fuzzy edges,leading to poor accuracy and missed detections.In order to resolve these problems,we propose MSCM-Net(Multi-Scale Cross-Modal Network),a multiscale cross-modal framework focused on detecting rail surface defects.MSCM-Net introduces an attention mechanism to dynamically weight the fusion of RGB and depth maps,effectively capturing and enhancing features at different scales for each modality.To further enrich feature representation and improve edge detection in blurred areas,we propose a multi-scale void fusion module that integrates multi-scale feature information.To improve cross-modal feature fusion,we develop a cross-enhanced fusion module that transfers fused features between layers to incorporate interlayer information.We also introduce a multimodal feature integration module,which merges modality-specific features from separate decoders into a shared decoder,enhancing detection by leveraging richer complementary information.Finally,we validate MSCM-Net on the NEU RSDDS-AUG RGB-depth dataset,comparing it against 12 leading methods,and the results show that MSCM-Net achieves superior performance on all metrics.展开更多
With the rapid growth of socialmedia,the spread of fake news has become a growing problem,misleading the public and causing significant harm.As social media content is often composed of both images and text,the use of...With the rapid growth of socialmedia,the spread of fake news has become a growing problem,misleading the public and causing significant harm.As social media content is often composed of both images and text,the use of multimodal approaches for fake news detection has gained significant attention.To solve the problems existing in previous multi-modal fake news detection algorithms,such as insufficient feature extraction and insufficient use of semantic relations between modes,this paper proposes the MFFFND-Co(Multimodal Feature Fusion Fake News Detection with Co-Attention Block)model.First,the model deeply explores the textual content,image content,and frequency domain features.Then,it employs a Co-Attention mechanism for cross-modal fusion.Additionally,a semantic consistency detectionmodule is designed to quantify semantic deviations,thereby enhancing the performance of fake news detection.Experimentally verified on two commonly used datasets,Twitter and Weibo,the model achieved F1 scores of 90.0% and 94.0%,respectively,significantly outperforming the pre-modified MFFFND(Multimodal Feature Fusion Fake News Detection with Attention Block)model and surpassing other baseline models.This improves the accuracy of detecting fake information in artificial intelligence detection and engineering software detection.展开更多
Multimodal sentiment analysis aims to understand people’s emotions and opinions from diverse data.Concate-nating or multiplying various modalities is a traditional multi-modal sentiment analysis fusion method.This fu...Multimodal sentiment analysis aims to understand people’s emotions and opinions from diverse data.Concate-nating or multiplying various modalities is a traditional multi-modal sentiment analysis fusion method.This fusion method does not utilize the correlation information between modalities.To solve this problem,this paper proposes amodel based on amulti-head attention mechanism.First,after preprocessing the original data.Then,the feature representation is converted into a sequence of word vectors and positional encoding is introduced to better understand the semantic and sequential information in the input sequence.Next,the input coding sequence is fed into the transformer model for further processing and learning.At the transformer layer,a cross-modal attention consisting of a pair of multi-head attention modules is employed to reflect the correlation between modalities.Finally,the processed results are input into the feedforward neural network to obtain the emotional output through the classification layer.Through the above processing flow,the model can capture semantic information and contextual relationships and achieve good results in various natural language processing tasks.Our model was tested on the CMU Multimodal Opinion Sentiment and Emotion Intensity(CMU-MOSEI)and Multimodal EmotionLines Dataset(MELD),achieving an accuracy of 82.04% and F1 parameters reached 80.59% on the former dataset.展开更多
With the explosive growth of false information on social media platforms, the automatic detection of multimodalfalse information has received increasing attention. Recent research has significantly contributed to mult...With the explosive growth of false information on social media platforms, the automatic detection of multimodalfalse information has received increasing attention. Recent research has significantly contributed to multimodalinformation exchange and fusion, with many methods attempting to integrate unimodal features to generatemultimodal news representations. However, they still need to fully explore the hierarchical and complex semanticcorrelations between different modal contents, severely limiting their performance detecting multimodal falseinformation. This work proposes a two-stage detection framework for multimodal false information detection,called ASMFD, which is based on image aesthetic similarity to segment and explores the consistency andinconsistency features of images and texts. Specifically, we first use the Contrastive Language-Image Pre-training(CLIP) model to learn the relationship between text and images through label awareness and train an imageaesthetic attribute scorer using an aesthetic attribute dataset. Then, we calculate the aesthetic similarity betweenthe image and related images and use this similarity as a threshold to divide the multimodal correlation matrixinto consistency and inconsistencymatrices. Finally, the fusionmodule is designed to identify essential features fordetectingmultimodal false information. In extensive experiments on four datasets, the performance of the ASMFDis superior to state-of-the-art baseline methods.展开更多
Iced transmission line galloping poses a significant threat to the safety and reliability of power systems,leading directly to line tripping,disconnections,and power outages.Existing early warning methods of iced tran...Iced transmission line galloping poses a significant threat to the safety and reliability of power systems,leading directly to line tripping,disconnections,and power outages.Existing early warning methods of iced transmission line galloping suffer from issues such as reliance on a single data source,neglect of irregular time series,and lack of attention-based closed-loop feedback,resulting in high rates of missed and false alarms.To address these challenges,we propose an Internet of Things(IoT)empowered early warning method of transmission line galloping that integrates time series data from optical fiber sensing and weather forecast.Initially,the method applies a primary adaptive weighted fusion to the IoT empowered optical fiber real-time sensing data and weather forecast data,followed by a secondary fusion based on a Back Propagation(BP)neural network,and uses the K-medoids algorithm for clustering the fused data.Furthermore,an adaptive irregular time series perception adjustment module is introduced into the traditional Gated Recurrent Unit(GRU)network,and closed-loop feedback based on attentionmechanism is employed to update network parameters through gradient feedback of the loss function,enabling closed-loop training and time series data prediction of the GRU network model.Subsequently,considering various types of prediction data and the duration of icing,an iced transmission line galloping risk coefficient is established,and warnings are categorized based on this coefficient.Finally,using an IoT-driven realistic dataset of iced transmission line galloping,the effectiveness of the proposed method is validated through multi-dimensional simulation scenarios.展开更多
Viral infectious diseases,characterized by their intricate nature and wide-ranging diversity,pose substantial challenges in the domain of data management.The vast volume of data generated by these diseases,spanning fr...Viral infectious diseases,characterized by their intricate nature and wide-ranging diversity,pose substantial challenges in the domain of data management.The vast volume of data generated by these diseases,spanning from the molecular mechanisms within cells to large-scale epidemiological patterns,has surpassed the capabilities of traditional analytical methods.In the era of artificial intelligence(AI)and big data,there is an urgent necessity for the optimization of these analytical methods to more effectively handle and utilize the information.Despite the rapid accumulation of data associated with viral infections,the lack of a comprehensive framework for integrating,selecting,and analyzing these datasets has left numerous researchers uncertain about which data to select,how to access it,and how to utilize it most effectively in their research.This review endeavors to fill these gaps by exploring the multifaceted nature of viral infectious diseases and summarizing relevant data across multiple levels,from the molecular details of pathogens to broad epidemiological trends.The scope extends from the micro-scale to the macro-scale,encompassing pathogens,hosts,and vectors.In addition to data summarization,this review thoroughly investigates various dataset sources.It also traces the historical evolution of data collection in the field of viral infectious diseases,highlighting the progress achieved over time.Simultaneously,it evaluates the current limitations that impede data utilization.Furthermore,we propose strategies to surmount these challenges,focusing on the development and application of advanced computational techniques,AI-driven models,and enhanced data integration practices.By providing a comprehensive synthesis of existing knowledge,this review is designed to guide future research and contribute to more informed approaches in the surveillance,prevention,and control of viral infectious diseases,particularly within the context of the expanding big-data landscape.展开更多
The Intelligent Internet of Things(IIoT)involves real-world things that communicate or interact with each other through networking technologies by collecting data from these“things”and using intelligent approaches,s...The Intelligent Internet of Things(IIoT)involves real-world things that communicate or interact with each other through networking technologies by collecting data from these“things”and using intelligent approaches,such as Artificial Intelligence(AI)and machine learning,to make accurate decisions.Data science is the science of dealing with data and its relationships through intelligent approaches.Most state-of-the-art research focuses independently on either data science or IIoT,rather than exploring their integration.Therefore,to address the gap,this article provides a comprehensive survey on the advances and integration of data science with the Intelligent IoT(IIoT)system by classifying the existing IoT-based data science techniques and presenting a summary of various characteristics.The paper analyzes the data science or big data security and privacy features,including network architecture,data protection,and continuous monitoring of data,which face challenges in various IoT-based systems.Extensive insights into IoT data security,privacy,and challenges are visualized in the context of data science for IoT.In addition,this study reveals the current opportunities to enhance data science and IoT market development.The current gap and challenges faced in the integration of data science and IoT are comprehensively presented,followed by the future outlook and possible solutions.展开更多
Air pollution in China covers a large area with complex sources and formation mechanisms,making it a unique place to conduct air pollution and atmospheric chemistry research.The National Natural Science Foundation of ...Air pollution in China covers a large area with complex sources and formation mechanisms,making it a unique place to conduct air pollution and atmospheric chemistry research.The National Natural Science Foundation of China’s Major Research Plan entitled“Fundamental Researches on the Formation and Response Mechanism of the Air Pollution Complex in China”(or the Plan)has funded 76 research projects to explore the causes of air pollution in China,and the key processes of air pollution in atmospheric physics and atmospheric chemistry.In order to summarize the abundant data from the Plan and exhibit the long-term impacts domestically and internationally,an integration project is responsible for collecting the various types of data generated by the 76 projects of the Plan.This project has classified and integrated these data,forming eight categories containing 258 datasets and 15 technical reports in total.The integration project has led to the successful establishment of the China Air Pollution Data Center(CAPDC)platform,providing storage,retrieval,and download services for the eight categories.This platform has distinct features including data visualization,related project information querying,and bilingual services in both English and Chinese,which allows for rapid searching and downloading of data and provides a solid foundation of data and support for future related research.Air pollution control in China,especially in the past decade,is undeniably a global exemplar,and this data center is the first in China to focus on research into the country’s air pollution complex.展开更多
As a new type of production factor in healthcare,healthcare data elements have been rapidly integrated into various health production processes,such as clinical assistance,health management,biological testing,and oper...As a new type of production factor in healthcare,healthcare data elements have been rapidly integrated into various health production processes,such as clinical assistance,health management,biological testing,and operation and supervision[1,2].Healthcare data elements include biolog.ical and clinical data that are related to disease,environ-mental health data that are associated with life,and operational and healthcare management data that are related to healthcare activities(Figure 1).Activities such as the construction of a data value assessment system,the devel-opment of a data circulation and sharing platform,and the authorization of data compliance and operation products support the strong growth momentum of the market for health care data elements in China[3].展开更多
As smart grid technology rapidly advances,the vast amount of user data collected by smart meter presents significant challenges in data security and privacy protection.Current research emphasizes data security and use...As smart grid technology rapidly advances,the vast amount of user data collected by smart meter presents significant challenges in data security and privacy protection.Current research emphasizes data security and user privacy concerns within smart grids.However,existing methods struggle with efficiency and security when processing large-scale data.Balancing efficient data processing with stringent privacy protection during data aggregation in smart grids remains an urgent challenge.This paper proposes an AI-based multi-type data aggregation method designed to enhance aggregation efficiency and security by standardizing and normalizing various data modalities.The approach optimizes data preprocessing,integrates Long Short-Term Memory(LSTM)networks for handling time-series data,and employs homomorphic encryption to safeguard user privacy.It also explores the application of Boneh Lynn Shacham(BLS)signatures for user authentication.The proposed scheme’s efficiency,security,and privacy protection capabilities are validated through rigorous security proofs and experimental analysis.展开更多
Earth’s internal core and crustal magnetic fields,as measured by geomagnetic satellites like MSS-1(Macao Science Satellite-1)and Swarm,are vital for understanding core dynamics and tectonic evolution.To model these i...Earth’s internal core and crustal magnetic fields,as measured by geomagnetic satellites like MSS-1(Macao Science Satellite-1)and Swarm,are vital for understanding core dynamics and tectonic evolution.To model these internal magnetic fields accurately,data selection based on specific criteria is often employed to minimize the influence of rapidly changing current systems in the ionosphere and magnetosphere.However,the quantitative impact of various data selection criteria on internal geomagnetic field modeling is not well understood.This study aims to address this issue and provide a reference for constructing and applying geomagnetic field models.First,we collect the latest MSS-1 and Swarm satellite magnetic data and summarize widely used data selection criteria in geomagnetic field modeling.Second,we briefly describe the method to co-estimate the core,crustal,and large-scale magnetospheric fields using satellite magnetic data.Finally,we conduct a series of field modeling experiments with different data selection criteria to quantitatively estimate their influence.Our numerical experiments confirm that without selecting data from dark regions and geomagnetically quiet times,the resulting internal field differences at the Earth’s surface can range from tens to hundreds of nanotesla(nT).Additionally,we find that the uncertainties introduced into field models by different data selection criteria are significantly larger than the measurement accuracy of modern geomagnetic satellites.These uncertainties should be considered when utilizing constructed magnetic field models for scientific research and applications.展开更多
Many fields,such as neuroscience,are experiencing the vast prolife ration of cellular data,underscoring the need fo r organizing and interpreting large datasets.A popular approach partitions data into manageable subse...Many fields,such as neuroscience,are experiencing the vast prolife ration of cellular data,underscoring the need fo r organizing and interpreting large datasets.A popular approach partitions data into manageable subsets via hierarchical clustering,but objective methods to determine the appropriate classification granularity are missing.We recently introduced a technique to systematically identify when to stop subdividing clusters based on the fundamental principle that cells must differ more between than within clusters.Here we present the corresponding protocol to classify cellular datasets by combining datadriven unsupervised hierarchical clustering with statistical testing.These general-purpose functions are applicable to any cellular dataset that can be organized as two-dimensional matrices of numerical values,including molecula r,physiological,and anatomical datasets.We demonstrate the protocol using cellular data from the Janelia MouseLight project to chara cterize morphological aspects of neurons.展开更多
On October 18,2017,the 19th National Congress Report called for the implementation of the Healthy China Strategy.The development of biomedical data plays a pivotal role in advancing this strategy.Since the 18th Nation...On October 18,2017,the 19th National Congress Report called for the implementation of the Healthy China Strategy.The development of biomedical data plays a pivotal role in advancing this strategy.Since the 18th National Congress of the Communist Party of China,China has vigorously promoted the integration and implementation of the Healthy China and Digital China strategies.The National Health Commission has prioritized the development of health and medical big data,issuing policies to promote standardized applica-tions and foster innovation in"Internet+Healthcare."Biomedical data has significantly contributed to preci-sion medicine,personalized health management,drug development,disease diagnosis,public health monitor-ing,and epidemic prediction capabilities.展开更多
The security of the seed industry is crucial for ensuring national food security.Currently,developed countries in Europe and America,along with international seed industry giants,have entered the Breeding 4.0 era.This...The security of the seed industry is crucial for ensuring national food security.Currently,developed countries in Europe and America,along with international seed industry giants,have entered the Breeding 4.0 era.This era integrates biotechnology,artificial intelligence(AI),and big data information technology.In contrast,China is still in a transition period between stages 2.0 and 3.0,which primarily relies on conventional selection and molecular breeding.In the context of increasingly complex international situations,accurately identifying core issues in China's seed industry innovation and seizing the frontier of international seed technology are strategically important.These efforts are essential for ensuring food security and revitalizing the seed industry.This paper systematically analyzes the characteristics of crop breeding data from artificial selection to intelligent design breeding.It explores the applications and development trends of AI and big data in modern crop breeding from several key perspectives.These include highthroughput phenotype acquisition and analysis,multiomics big data database and management system construction,AI-based multiomics integrated analysis,and the development of intelligent breeding software tools based on biological big data and AI technology.Based on an in-depth analysis of the current status and challenges of China's seed industry technology development,we propose strategic goals and key tasks for China's new generation of AI and big data-driven intelligent design breeding.These suggestions aim to accelerate the development of an intelligent-driven crop breeding engineering system that features large-scale gene mining,efficient gene manipulation,engineered variety design,and systematized biobreeding.This study provides a theoretical basis and practical guidance for the development of China's seed industry technology.展开更多
The accelerated advancement of the Internet of Things(IoT)has generated substantial data,including sensitive and private information.Consequently,it is imperative to guarantee the security of data sharing.While facili...The accelerated advancement of the Internet of Things(IoT)has generated substantial data,including sensitive and private information.Consequently,it is imperative to guarantee the security of data sharing.While facilitating fine-grained access control,Ciphertext Policy Attribute-Based Encryption(CP-ABE)can effectively ensure the confidentiality of shared data.Nevertheless,the conventional centralized CP-ABE scheme is plagued by the issues of keymisuse,key escrow,and large computation,which will result in security risks.This paper suggests a lightweight IoT data security sharing scheme that integrates blockchain technology and CP-ABE to address the abovementioned issues.The integrity and traceability of shared data are guaranteed by the use of blockchain technology to store and verify access transactions.The encryption and decryption operations of the CP-ABE algorithm have been implemented using elliptic curve scalarmultiplication to accommodate lightweight IoT devices,as opposed to themore arithmetic bilinear pairing found in the traditional CP-ABE algorithm.Additionally,a portion of the computation is delegated to the edge nodes to alleviate the computational burden on users.A distributed key management method is proposed to address the issues of key escrow andmisuse.Thismethod employs the edge blockchain to facilitate the storage and distribution of attribute private keys.Meanwhile,data security sharing is enhanced by combining off-chain and on-chain ciphertext storage.The security and performance analysis indicates that the proposed scheme is more efficient and secure.展开更多
调查和分析元数据标准在健康科学数据中的应用现状,有助于为我国健康科学数据描述中元数据标准的选择、健康科学数据平台的建设提供参考。通过网络调研法对科学数据仓储注册系统(registry of research data repositories,re3data)中的...调查和分析元数据标准在健康科学数据中的应用现状,有助于为我国健康科学数据描述中元数据标准的选择、健康科学数据平台的建设提供参考。通过网络调研法对科学数据仓储注册系统(registry of research data repositories,re3data)中的健康科学数据管理平台进行调研,梳理所应用的元数据标准,分析典型元数据标准在平台中的应用情况,并归纳其在健康科学数据描述中的适用性。re3data中各健康科学数据平台共使用14种元数据标准,其中DC、DataCite、DDI、仓储自建元数据标准的使用最为广泛,多数平台组合使用多种元数据标准。各类元数据标准可分为通用型、社会科学型、自建型3类,分别适用于描述健康科学数据通用属性、社会科学研究产生的健康科学数据、特色和专业性强及政府开放的健康科学数据。展开更多
Population growth leads to increased utilization of water resources.One of these resources is groundwater,which has steadily declined each year.The depletion of these resources brings about various environmental chall...Population growth leads to increased utilization of water resources.One of these resources is groundwater,which has steadily declined each year.The depletion of these resources brings about various environmental challenges.The present study aimed to explore the relationship between groundwater fluctuations and land subsidence in the Malayer Plain,Iran,focusing on quantifying subsidence resulting from groundwater extraction.Using Sentinel-1 satellite data(2014–2019)and monthly piezometric measurements(1996–2018),the analysis revealed an average deformation velocity of–6.3 cm yr–1,with accumulated subsidence of–32 cm over the 2014–2019 period.The maximum subsidence rate reached 10.3 cm yr–1 in areas of intensive agricultural activity.A wavelet-PCA spatiotemporal analysis of groundwater fluctuations identified critical multi-scale patterns strongly correlated with subsidence trends.Regression analysis between subsidence rates and groundwater fluctuations at various wavelet decomposition levels explained 75%of the variance(R2=0.75),indicating that intermediate-scale groundwater declines were the primary drivers of subsidence.Furthermore,land use analysis using Landsat data(1999–2021)revealed a 6230-ha increase in irrigated farmland,contributing to heightened groundwater extraction and subsidence rates.These findings highlight the critical need for sustainable groundwater management to mitigate the risks of continued subsidence in the region.展开更多
This study presents a machine learning-based method for predicting fragment velocity distribution in warhead fragmentation under explosive loading condition.The fragment resultant velocities are correlated with key de...This study presents a machine learning-based method for predicting fragment velocity distribution in warhead fragmentation under explosive loading condition.The fragment resultant velocities are correlated with key design parameters including casing dimensions and detonation positions.The paper details the finite element analysis for fragmentation,the characterizations of the dynamic hardening and fracture models,the generation of comprehensive datasets,and the training of the ANN model.The results show the influence of casing dimensions on fragment velocity distributions,with the tendencies indicating increased resultant velocity with reduced thickness,increased length and diameter.The model's predictive capability is demonstrated through the accurate predictions for both training and testing datasets,showing its potential for the real-time prediction of fragmentation performance.展开更多
基金funded by the Scientific Funding for China Academy of Railway Sciences Corporation Limited,China(No.2023YJ125).
文摘Speech-face association aims to achieve identity matching between facial images and voice segments by aligning cross-modal features.Existing research primarily focuses on learning shared-space representations and computing one-to-one similarities between cross-modal sample pairs to establish their correlation.However,these approaches do not fully account for intra-class variations between the modalities or the many-to-many relationships among cross-modal samples,which are crucial for robust association modeling.To address these challenges,we propose a novel framework that leverages global information to align voice and face embeddings while effectively correlating identity information embedded in both modalities.First,we jointly pre-train face recognition and speaker recognition networks to encode discriminative features from facial images and voice segments.This shared pre-training step ensures the extraction of complementary identity information across modalities.Subsequently,we introduce a cross-modal simplex center loss,which aligns samples with identity centers located at the vertices of a regular simplex inscribed on a hypersphere.This design enforces an equidistant and balanced distribution of identity embeddings,reducing intra-class variations.Furthermore,we employ an improved triplet center loss that emphasizes hard sample mining and optimizes inter-class separability,enhancing the model’s ability to generalize across challenging scenarios.Extensive experiments validate the effectiveness of our framework,demonstrating superior performance across various speech-face association tasks,including matching,verification,and retrieval.Notably,in the challenging gender-constrained matching task,our method achieves a remarkable accuracy of 79.22%,significantly outperforming existing approaches.These results highlight the potential of the proposed framework to advance the state of the art in cross-modal identity association.
基金funded by the Institute of Information&CommunicationsTechnology Planning&Evaluation(IITP)grant funded by the Korea government(MSIT),grant number 2021-0-01341.
文摘Emotion recognition under uncontrolled and noisy environments presents persistent challenges in the design of emotionally responsive systems.The current study introduces an audio-visual recognition framework designed to address performance degradation caused by environmental interference,such as background noise,overlapping speech,and visual obstructions.The proposed framework employs a structured fusion approach,combining early-stage feature-level integration with decision-level coordination guided by temporal attention mechanisms.Audio data are transformed into mel-spectrogram representations,and visual data are represented as raw frame sequences.Spatial and temporal features are extracted through convolutional and transformer-based encoders,allowing the framework to capture complementary and hierarchical information fromboth sources.Across-modal attentionmodule enables selective emphasis on relevant signals while suppressing modality-specific noise.Performance is validated on a modified version of the AFEW dataset,in which controlled noise is introduced to emulate realistic conditions.The framework achieves higher classification accuracy than comparative baselines,confirming increased robustness under conditions of cross-modal disruption.This result demonstrates the suitability of the proposed method for deployment in practical emotion-aware technologies operating outside controlled environments.The study also contributes a systematic approach to fusion design and supports further exploration in the direction of resilientmultimodal emotion analysis frameworks.The source code is publicly available at https://github.com/asmoon002/AVER(accessed on 18 August 2025).
基金funded by the National Natural Science Foundation of China(grant number 62306186)the Technology Plan Joint Foundation of Liaoning Province(grant number 2023-MSLH-246)the Technology Plan Joint Foundation of Liaoning Province(grant number 2023-BSBA-238).
文摘Detecting surface defects on unused rails is crucial for evaluating rail quality and durability to ensure the safety of rail transportation.However,existing detection methods often struggle with challenges such as complex defect morphology,texture similarity,and fuzzy edges,leading to poor accuracy and missed detections.In order to resolve these problems,we propose MSCM-Net(Multi-Scale Cross-Modal Network),a multiscale cross-modal framework focused on detecting rail surface defects.MSCM-Net introduces an attention mechanism to dynamically weight the fusion of RGB and depth maps,effectively capturing and enhancing features at different scales for each modality.To further enrich feature representation and improve edge detection in blurred areas,we propose a multi-scale void fusion module that integrates multi-scale feature information.To improve cross-modal feature fusion,we develop a cross-enhanced fusion module that transfers fused features between layers to incorporate interlayer information.We also introduce a multimodal feature integration module,which merges modality-specific features from separate decoders into a shared decoder,enhancing detection by leveraging richer complementary information.Finally,we validate MSCM-Net on the NEU RSDDS-AUG RGB-depth dataset,comparing it against 12 leading methods,and the results show that MSCM-Net achieves superior performance on all metrics.
基金supported by Communication University of China(HG23035)partly supported by the Fundamental Research Funds for the Central Universities(CUC230A013).
文摘With the rapid growth of socialmedia,the spread of fake news has become a growing problem,misleading the public and causing significant harm.As social media content is often composed of both images and text,the use of multimodal approaches for fake news detection has gained significant attention.To solve the problems existing in previous multi-modal fake news detection algorithms,such as insufficient feature extraction and insufficient use of semantic relations between modes,this paper proposes the MFFFND-Co(Multimodal Feature Fusion Fake News Detection with Co-Attention Block)model.First,the model deeply explores the textual content,image content,and frequency domain features.Then,it employs a Co-Attention mechanism for cross-modal fusion.Additionally,a semantic consistency detectionmodule is designed to quantify semantic deviations,thereby enhancing the performance of fake news detection.Experimentally verified on two commonly used datasets,Twitter and Weibo,the model achieved F1 scores of 90.0% and 94.0%,respectively,significantly outperforming the pre-modified MFFFND(Multimodal Feature Fusion Fake News Detection with Attention Block)model and surpassing other baseline models.This improves the accuracy of detecting fake information in artificial intelligence detection and engineering software detection.
基金supported by the National Natural Science Foundation of China under Grant 61702462the Henan Provincial Science and Technology Research Project under Grants 222102210010 and 222102210064+2 种基金the Research and Practice Project of Higher Education Teaching Reform in Henan Province under Grants 2019SJGLX320 and 2019SJGLX020the Undergraduate Universities Smart Teaching Special Research Project of Henan Province under Grant JiaoGao[2021]No.489-29the Academic Degrees&Graduate Education Reform Project of Henan Province under Grant 2021SJGLX115Y.
文摘Multimodal sentiment analysis aims to understand people’s emotions and opinions from diverse data.Concate-nating or multiplying various modalities is a traditional multi-modal sentiment analysis fusion method.This fusion method does not utilize the correlation information between modalities.To solve this problem,this paper proposes amodel based on amulti-head attention mechanism.First,after preprocessing the original data.Then,the feature representation is converted into a sequence of word vectors and positional encoding is introduced to better understand the semantic and sequential information in the input sequence.Next,the input coding sequence is fed into the transformer model for further processing and learning.At the transformer layer,a cross-modal attention consisting of a pair of multi-head attention modules is employed to reflect the correlation between modalities.Finally,the processed results are input into the feedforward neural network to obtain the emotional output through the classification layer.Through the above processing flow,the model can capture semantic information and contextual relationships and achieve good results in various natural language processing tasks.Our model was tested on the CMU Multimodal Opinion Sentiment and Emotion Intensity(CMU-MOSEI)and Multimodal EmotionLines Dataset(MELD),achieving an accuracy of 82.04% and F1 parameters reached 80.59% on the former dataset.
文摘With the explosive growth of false information on social media platforms, the automatic detection of multimodalfalse information has received increasing attention. Recent research has significantly contributed to multimodalinformation exchange and fusion, with many methods attempting to integrate unimodal features to generatemultimodal news representations. However, they still need to fully explore the hierarchical and complex semanticcorrelations between different modal contents, severely limiting their performance detecting multimodal falseinformation. This work proposes a two-stage detection framework for multimodal false information detection,called ASMFD, which is based on image aesthetic similarity to segment and explores the consistency andinconsistency features of images and texts. Specifically, we first use the Contrastive Language-Image Pre-training(CLIP) model to learn the relationship between text and images through label awareness and train an imageaesthetic attribute scorer using an aesthetic attribute dataset. Then, we calculate the aesthetic similarity betweenthe image and related images and use this similarity as a threshold to divide the multimodal correlation matrixinto consistency and inconsistencymatrices. Finally, the fusionmodule is designed to identify essential features fordetectingmultimodal false information. In extensive experiments on four datasets, the performance of the ASMFDis superior to state-of-the-art baseline methods.
基金research was funded by Science and Technology Project of State Grid Corporation of China under grant number 5200-202319382A-2-3-XG.
文摘Iced transmission line galloping poses a significant threat to the safety and reliability of power systems,leading directly to line tripping,disconnections,and power outages.Existing early warning methods of iced transmission line galloping suffer from issues such as reliance on a single data source,neglect of irregular time series,and lack of attention-based closed-loop feedback,resulting in high rates of missed and false alarms.To address these challenges,we propose an Internet of Things(IoT)empowered early warning method of transmission line galloping that integrates time series data from optical fiber sensing and weather forecast.Initially,the method applies a primary adaptive weighted fusion to the IoT empowered optical fiber real-time sensing data and weather forecast data,followed by a secondary fusion based on a Back Propagation(BP)neural network,and uses the K-medoids algorithm for clustering the fused data.Furthermore,an adaptive irregular time series perception adjustment module is introduced into the traditional Gated Recurrent Unit(GRU)network,and closed-loop feedback based on attentionmechanism is employed to update network parameters through gradient feedback of the loss function,enabling closed-loop training and time series data prediction of the GRU network model.Subsequently,considering various types of prediction data and the duration of icing,an iced transmission line galloping risk coefficient is established,and warnings are categorized based on this coefficient.Finally,using an IoT-driven realistic dataset of iced transmission line galloping,the effectiveness of the proposed method is validated through multi-dimensional simulation scenarios.
基金supported by the National Natural Science Foundation of China(32370703)the CAMS Innovation Fund for Medical Sciences(CIFMS)(2022-I2M-1-021,2021-I2M-1-061)the Major Project of Guangzhou National Labora-tory(GZNL2024A01015).
文摘Viral infectious diseases,characterized by their intricate nature and wide-ranging diversity,pose substantial challenges in the domain of data management.The vast volume of data generated by these diseases,spanning from the molecular mechanisms within cells to large-scale epidemiological patterns,has surpassed the capabilities of traditional analytical methods.In the era of artificial intelligence(AI)and big data,there is an urgent necessity for the optimization of these analytical methods to more effectively handle and utilize the information.Despite the rapid accumulation of data associated with viral infections,the lack of a comprehensive framework for integrating,selecting,and analyzing these datasets has left numerous researchers uncertain about which data to select,how to access it,and how to utilize it most effectively in their research.This review endeavors to fill these gaps by exploring the multifaceted nature of viral infectious diseases and summarizing relevant data across multiple levels,from the molecular details of pathogens to broad epidemiological trends.The scope extends from the micro-scale to the macro-scale,encompassing pathogens,hosts,and vectors.In addition to data summarization,this review thoroughly investigates various dataset sources.It also traces the historical evolution of data collection in the field of viral infectious diseases,highlighting the progress achieved over time.Simultaneously,it evaluates the current limitations that impede data utilization.Furthermore,we propose strategies to surmount these challenges,focusing on the development and application of advanced computational techniques,AI-driven models,and enhanced data integration practices.By providing a comprehensive synthesis of existing knowledge,this review is designed to guide future research and contribute to more informed approaches in the surveillance,prevention,and control of viral infectious diseases,particularly within the context of the expanding big-data landscape.
基金supported in part by the National Natural Science Foundation of China under Grant 62371181in part by the Changzhou Science and Technology International Cooperation Program under Grant CZ20230029+1 种基金supported by a National Research Foundation of Korea(NRF)grant funded by the Korea government(MSIT)(2021R1A2B5B02087169)supported under the framework of international cooperation program managed by the National Research Foundation of Korea(2022K2A9A1A01098051)。
文摘The Intelligent Internet of Things(IIoT)involves real-world things that communicate or interact with each other through networking technologies by collecting data from these“things”and using intelligent approaches,such as Artificial Intelligence(AI)and machine learning,to make accurate decisions.Data science is the science of dealing with data and its relationships through intelligent approaches.Most state-of-the-art research focuses independently on either data science or IIoT,rather than exploring their integration.Therefore,to address the gap,this article provides a comprehensive survey on the advances and integration of data science with the Intelligent IoT(IIoT)system by classifying the existing IoT-based data science techniques and presenting a summary of various characteristics.The paper analyzes the data science or big data security and privacy features,including network architecture,data protection,and continuous monitoring of data,which face challenges in various IoT-based systems.Extensive insights into IoT data security,privacy,and challenges are visualized in the context of data science for IoT.In addition,this study reveals the current opportunities to enhance data science and IoT market development.The current gap and challenges faced in the integration of data science and IoT are comprehensively presented,followed by the future outlook and possible solutions.
基金supported by the National Natural Science Foundation of China(Grant No.92044303)。
文摘Air pollution in China covers a large area with complex sources and formation mechanisms,making it a unique place to conduct air pollution and atmospheric chemistry research.The National Natural Science Foundation of China’s Major Research Plan entitled“Fundamental Researches on the Formation and Response Mechanism of the Air Pollution Complex in China”(or the Plan)has funded 76 research projects to explore the causes of air pollution in China,and the key processes of air pollution in atmospheric physics and atmospheric chemistry.In order to summarize the abundant data from the Plan and exhibit the long-term impacts domestically and internationally,an integration project is responsible for collecting the various types of data generated by the 76 projects of the Plan.This project has classified and integrated these data,forming eight categories containing 258 datasets and 15 technical reports in total.The integration project has led to the successful establishment of the China Air Pollution Data Center(CAPDC)platform,providing storage,retrieval,and download services for the eight categories.This platform has distinct features including data visualization,related project information querying,and bilingual services in both English and Chinese,which allows for rapid searching and downloading of data and provides a solid foundation of data and support for future related research.Air pollution control in China,especially in the past decade,is undeniably a global exemplar,and this data center is the first in China to focus on research into the country’s air pollution complex.
基金supported by National Natural Science Foundation of China(Grants 72474022,71974011,72174022,71972012,71874009)"BIT think tank"Promotion Plan of Science and Technology Innovation Program of Beijing Institute of Technology(Grants 2024CX14017,2023CX13029).
文摘As a new type of production factor in healthcare,healthcare data elements have been rapidly integrated into various health production processes,such as clinical assistance,health management,biological testing,and operation and supervision[1,2].Healthcare data elements include biolog.ical and clinical data that are related to disease,environ-mental health data that are associated with life,and operational and healthcare management data that are related to healthcare activities(Figure 1).Activities such as the construction of a data value assessment system,the devel-opment of a data circulation and sharing platform,and the authorization of data compliance and operation products support the strong growth momentum of the market for health care data elements in China[3].
基金supported by the National Key R&D Program of China(No.2023YFB2703700)the National Natural Science Foundation of China(Nos.U21A20465,62302457,62402444,62172292)+4 种基金the Fundamental Research Funds of Zhejiang Sci-Tech University(Nos.23222092-Y,22222266-Y)the Program for Leading Innovative Research Team of Zhejiang Province(No.2023R01001)the Zhejiang Provincial Natural Science Foundation of China(Nos.LQ24F020008,LQ24F020012)the Foundation of State Key Laboratory of Public Big Data(No.[2022]417)the“Pioneer”and“Leading Goose”R&D Program of Zhejiang(No.2023C01119).
文摘As smart grid technology rapidly advances,the vast amount of user data collected by smart meter presents significant challenges in data security and privacy protection.Current research emphasizes data security and user privacy concerns within smart grids.However,existing methods struggle with efficiency and security when processing large-scale data.Balancing efficient data processing with stringent privacy protection during data aggregation in smart grids remains an urgent challenge.This paper proposes an AI-based multi-type data aggregation method designed to enhance aggregation efficiency and security by standardizing and normalizing various data modalities.The approach optimizes data preprocessing,integrates Long Short-Term Memory(LSTM)networks for handling time-series data,and employs homomorphic encryption to safeguard user privacy.It also explores the application of Boneh Lynn Shacham(BLS)signatures for user authentication.The proposed scheme’s efficiency,security,and privacy protection capabilities are validated through rigorous security proofs and experimental analysis.
基金supported by the National Natural Science Foundation of China(42250101)the Macao Foundation。
文摘Earth’s internal core and crustal magnetic fields,as measured by geomagnetic satellites like MSS-1(Macao Science Satellite-1)and Swarm,are vital for understanding core dynamics and tectonic evolution.To model these internal magnetic fields accurately,data selection based on specific criteria is often employed to minimize the influence of rapidly changing current systems in the ionosphere and magnetosphere.However,the quantitative impact of various data selection criteria on internal geomagnetic field modeling is not well understood.This study aims to address this issue and provide a reference for constructing and applying geomagnetic field models.First,we collect the latest MSS-1 and Swarm satellite magnetic data and summarize widely used data selection criteria in geomagnetic field modeling.Second,we briefly describe the method to co-estimate the core,crustal,and large-scale magnetospheric fields using satellite magnetic data.Finally,we conduct a series of field modeling experiments with different data selection criteria to quantitatively estimate their influence.Our numerical experiments confirm that without selecting data from dark regions and geomagnetically quiet times,the resulting internal field differences at the Earth’s surface can range from tens to hundreds of nanotesla(nT).Additionally,we find that the uncertainties introduced into field models by different data selection criteria are significantly larger than the measurement accuracy of modern geomagnetic satellites.These uncertainties should be considered when utilizing constructed magnetic field models for scientific research and applications.
基金supported in part by NIH grants R01NS39600,U01MH114829RF1MH128693(to GAA)。
文摘Many fields,such as neuroscience,are experiencing the vast prolife ration of cellular data,underscoring the need fo r organizing and interpreting large datasets.A popular approach partitions data into manageable subsets via hierarchical clustering,but objective methods to determine the appropriate classification granularity are missing.We recently introduced a technique to systematically identify when to stop subdividing clusters based on the fundamental principle that cells must differ more between than within clusters.Here we present the corresponding protocol to classify cellular datasets by combining datadriven unsupervised hierarchical clustering with statistical testing.These general-purpose functions are applicable to any cellular dataset that can be organized as two-dimensional matrices of numerical values,including molecula r,physiological,and anatomical datasets.We demonstrate the protocol using cellular data from the Janelia MouseLight project to chara cterize morphological aspects of neurons.
文摘On October 18,2017,the 19th National Congress Report called for the implementation of the Healthy China Strategy.The development of biomedical data plays a pivotal role in advancing this strategy.Since the 18th National Congress of the Communist Party of China,China has vigorously promoted the integration and implementation of the Healthy China and Digital China strategies.The National Health Commission has prioritized the development of health and medical big data,issuing policies to promote standardized applica-tions and foster innovation in"Internet+Healthcare."Biomedical data has significantly contributed to preci-sion medicine,personalized health management,drug development,disease diagnosis,public health monitor-ing,and epidemic prediction capabilities.
基金partially supported by the Construction of Collaborative Innovation Center of Beijing Academy of Agricultural and Forestry Sciences(KJCX20240406)the Beijing Natural Science Foundation(JQ24037)+1 种基金the National Natural Science Foundation of China(32330075)the Earmarked Fund for China Agriculture Research System(CARS-02 and CARS-54)。
文摘The security of the seed industry is crucial for ensuring national food security.Currently,developed countries in Europe and America,along with international seed industry giants,have entered the Breeding 4.0 era.This era integrates biotechnology,artificial intelligence(AI),and big data information technology.In contrast,China is still in a transition period between stages 2.0 and 3.0,which primarily relies on conventional selection and molecular breeding.In the context of increasingly complex international situations,accurately identifying core issues in China's seed industry innovation and seizing the frontier of international seed technology are strategically important.These efforts are essential for ensuring food security and revitalizing the seed industry.This paper systematically analyzes the characteristics of crop breeding data from artificial selection to intelligent design breeding.It explores the applications and development trends of AI and big data in modern crop breeding from several key perspectives.These include highthroughput phenotype acquisition and analysis,multiomics big data database and management system construction,AI-based multiomics integrated analysis,and the development of intelligent breeding software tools based on biological big data and AI technology.Based on an in-depth analysis of the current status and challenges of China's seed industry technology development,we propose strategic goals and key tasks for China's new generation of AI and big data-driven intelligent design breeding.These suggestions aim to accelerate the development of an intelligent-driven crop breeding engineering system that features large-scale gene mining,efficient gene manipulation,engineered variety design,and systematized biobreeding.This study provides a theoretical basis and practical guidance for the development of China's seed industry technology.
文摘The accelerated advancement of the Internet of Things(IoT)has generated substantial data,including sensitive and private information.Consequently,it is imperative to guarantee the security of data sharing.While facilitating fine-grained access control,Ciphertext Policy Attribute-Based Encryption(CP-ABE)can effectively ensure the confidentiality of shared data.Nevertheless,the conventional centralized CP-ABE scheme is plagued by the issues of keymisuse,key escrow,and large computation,which will result in security risks.This paper suggests a lightweight IoT data security sharing scheme that integrates blockchain technology and CP-ABE to address the abovementioned issues.The integrity and traceability of shared data are guaranteed by the use of blockchain technology to store and verify access transactions.The encryption and decryption operations of the CP-ABE algorithm have been implemented using elliptic curve scalarmultiplication to accommodate lightweight IoT devices,as opposed to themore arithmetic bilinear pairing found in the traditional CP-ABE algorithm.Additionally,a portion of the computation is delegated to the edge nodes to alleviate the computational burden on users.A distributed key management method is proposed to address the issues of key escrow andmisuse.Thismethod employs the edge blockchain to facilitate the storage and distribution of attribute private keys.Meanwhile,data security sharing is enhanced by combining off-chain and on-chain ciphertext storage.The security and performance analysis indicates that the proposed scheme is more efficient and secure.
文摘调查和分析元数据标准在健康科学数据中的应用现状,有助于为我国健康科学数据描述中元数据标准的选择、健康科学数据平台的建设提供参考。通过网络调研法对科学数据仓储注册系统(registry of research data repositories,re3data)中的健康科学数据管理平台进行调研,梳理所应用的元数据标准,分析典型元数据标准在平台中的应用情况,并归纳其在健康科学数据描述中的适用性。re3data中各健康科学数据平台共使用14种元数据标准,其中DC、DataCite、DDI、仓储自建元数据标准的使用最为广泛,多数平台组合使用多种元数据标准。各类元数据标准可分为通用型、社会科学型、自建型3类,分别适用于描述健康科学数据通用属性、社会科学研究产生的健康科学数据、特色和专业性强及政府开放的健康科学数据。
文摘Population growth leads to increased utilization of water resources.One of these resources is groundwater,which has steadily declined each year.The depletion of these resources brings about various environmental challenges.The present study aimed to explore the relationship between groundwater fluctuations and land subsidence in the Malayer Plain,Iran,focusing on quantifying subsidence resulting from groundwater extraction.Using Sentinel-1 satellite data(2014–2019)and monthly piezometric measurements(1996–2018),the analysis revealed an average deformation velocity of–6.3 cm yr–1,with accumulated subsidence of–32 cm over the 2014–2019 period.The maximum subsidence rate reached 10.3 cm yr–1 in areas of intensive agricultural activity.A wavelet-PCA spatiotemporal analysis of groundwater fluctuations identified critical multi-scale patterns strongly correlated with subsidence trends.Regression analysis between subsidence rates and groundwater fluctuations at various wavelet decomposition levels explained 75%of the variance(R2=0.75),indicating that intermediate-scale groundwater declines were the primary drivers of subsidence.Furthermore,land use analysis using Landsat data(1999–2021)revealed a 6230-ha increase in irrigated farmland,contributing to heightened groundwater extraction and subsidence rates.These findings highlight the critical need for sustainable groundwater management to mitigate the risks of continued subsidence in the region.
基金supported by Poongsan-KAIST Future Research Center Projectthe fund support provided by the National Research Foundation of Korea(NRF)grant funded by the Korea government(MSIT)(Grant No.2023R1A2C2005661)。
文摘This study presents a machine learning-based method for predicting fragment velocity distribution in warhead fragmentation under explosive loading condition.The fragment resultant velocities are correlated with key design parameters including casing dimensions and detonation positions.The paper details the finite element analysis for fragmentation,the characterizations of the dynamic hardening and fracture models,the generation of comprehensive datasets,and the training of the ANN model.The results show the influence of casing dimensions on fragment velocity distributions,with the tendencies indicating increased resultant velocity with reduced thickness,increased length and diameter.The model's predictive capability is demonstrated through the accurate predictions for both training and testing datasets,showing its potential for the real-time prediction of fragmentation performance.
