Straw return increases crop production by improving soil organic carbon sequestration and soil aggregation in a long-term wheat-cotton cropping system

Straw return is a promising strategy for managing soil organic carbon(SOC)and improving yield stability.However,the optimal straw return strategy for sustainable crop production in the wheat(Triticum aestivum L.)-cotton(Gossypium hirsutum L.)cropping system remains uncertain.The objective of this study was to quantify the long-term(10 years)impact of carbon(C)input on SOC sequestration,soil aggregation and crop yields in a wheat-cotton cropping system in the Yangtze River Valley,China.Five treatments were arranged with a single-factor randomized design as follows:no straw return(Control),return of wheat straw only(Wt),return of cotton straw only(Ct),return of 50%wheat and 50%cotton straw(Wh-Ch)and return of 100%wheat and 100%cotton straw(Wt-Ct).In comparison to the Control,the SOC content increased by 8.4 to 20.2%under straw return.A significant linear positive correlation between SOC sequestration and C input(1.42-7.19 Mg ha^(−1)yr^(−1))(P<0.05)was detected.The percentages of aggregates of sizes>2 and 1-2 mm at the 0-20 cm soil depth were also significantly elevated under straw return,with the greatest increase of the aggregate stability in the Wt-Ct treatment(28.1%).The average wheat yields increased by 12.4-36.0%and cotton yields increased by 29.4-73.7%,and significantly linear positive correlations were also detected between C input and the yields of wheat and cotton.The average sustainable yield index(SYI)reached a maximum value of 0.69 when the C input was 7.08 Mg ha^(−1)yr^(−1),which was close to the maximum value(SYI of 0.69,C input of 7.19 Mg ha^(−1)yr^(-1))in the Wt-Ct treatment.Overall,the return of both wheat and cotton straw was the best strategy for improving SOC sequestration,soil aggregation,yields and their sustainability in the wheat-cotton rotation system.

A novel approach to Parkinson's disease treatment with a potentially dual-acting therapeutic agent that targetsα-synuclein aggregation and neuron death

Parkinson's disease(PD),a prevalent neurodegenerative disorder,is chara cterized by the loss of dopaminergic neurons and the aggregation ofα-synuclein protein into Lewy bodies.While the current standards of therapy have been successful in providing some symptom relief,they fail to address the underlying pathophysiology of PD and as a result,they have no effect on disease progression.

Impacts of Aggregation Methods and Trophospecies Number on the Structure and Function of Marine Food Webs

Aggregation of species with similar ecological properties is one of the effective methods to simplify food web researches.However,species aggregation will affect not only the complexity of modeling process but also the accuracy of models’outputs.Selection of aggregation methods and the number of trophospecies are the keys to study the simplification of food web.In this study,three aggregation methods,including taxonomic aggregation(TA),structural equivalence aggregation(SEA),and self-organizing maps(SOM),were analyzed and compared with the linear inverse model–Markov Chain Monte Carlo(LIM-MCMC)model.Impacts of aggregation methods and trophospecies number on food webs were evaluated based on the robustness and unitless of ecological net-work indices.Results showed that aggregation method of SEA performed better than the other two methods in estimating food web structure and function indices.The effects of aggregation methods were driven by the differences in species aggregation principles,which will alter food web structure and function through the redistribution of energy flow.According to the results of mean absolute percentage error(MAPE)which can be applied to evaluate the accuracy of the model,we found that MAPE in food web indices will increase with the reducing trophospecies number,and MAPE in food web function indices were smaller and more stable than those in food web structure indices.Therefore,trade-off between simplifying food webs and reflecting the status of ecosystem should be con-sidered in food web studies.These findings highlight the importance of aggregation methods and trophospecies number in the analy-sis of food web simplification.This study provided a framework to explore the extent to which food web models are affected by dif-ferent species aggregation,and will provide scientific basis for the construction of food webs.

Underwater Pulse Waveform Recognition Based on Hash Aggregate Discriminant Network

Underwater pulse waveform recognition is an important method for underwater object detection.Most existing works focus on the application of traditional pattern recognition methods,which ignore the time-and space-varying characteristics in sound propagation channels and cannot easily extract valuable waveform features.Sound propagation channels in seawater are time-and space-varying convolutional channels.In the extraction of the waveform features of underwater acoustic signals,the effect of high-accuracy underwater acoustic signal recognition is identified by eliminating the influence of time-and space-varying convolutional channels to the greatest extent possible.We propose a hash aggregate discriminative network(HADN),which combines hash learning and deep learning to minimize the time-and space-varying effects on convolutional channels and adaptively learns effective underwater waveform features to achieve high-accuracy underwater pulse waveform recognition.In the extraction of the hash features of acoustic signals,a discrete constraint between clusters within a hash feature class is introduced.This constraint can ensure that the influence of convolutional channels on hash features is minimized.In addition,we design a new loss function called aggregate discriminative loss(AD-loss).The use of AD-loss and softmax-loss can increase the discriminativeness of the learned hash features.Experimental results show that on pool and ocean datasets,which were collected in pools and oceans,respectively,by using acoustic collectors,the proposed HADN performs better than other comparative models in terms of accuracy and mAP.

Small molecules to target tau amyloid aggregation

Protein aggregation has been linked with many neurodegenerative diseases,such as Alzheimer’s disease(AD)or Parkinson’s disease.AD belongs to a group of heterogeneous and incurable neurodegenerative disorders collectively known as tauopathies.They comprise frontotemporal dementia,Pick’s disease,or corticobasal degeneration,among others.The symptomatology varies with the specific tau protein variant involved and the affected brain region or cell type.However,they share a common neuropathological hallmark-the formation of proteinaceous deposits named neurofibrillary tangles.Neurofibrillary tangles,primarily composed of aggregated tau(Zhang et al.,2022),disrupt normal neuronal functions,leading to cell death and cognitive decline.

Aggregation-regulated bioreduction process of graphene oxide by Shewanella bacteria

The bioreduction of graphene oxide(GO)using environmentally functional bacteria such as Shewanella represents a green approach to produce reduced graphene oxide(rGO).This process differs from the chemical reduction that involves instantaneous molecular reactions.In bioreduction,the contact of bacterial cells and GO is considered the rate-limiting step.To reveal how the bacteria-GO integration regulates rGO production,the comparative experiments of GO and three Shewanella strains were carried out.Fourier-transform infrared spectroscopy,X-ray photoelectron spectroscopy,Raman spectroscopy,and atomic force microscopy were used to characterize the reduction degree and the aggregation degree.The results showed that a spontaneous aggregation of GO and Shewanella into the condensed entity occurred within 36 h.A positive linear correlation was established,linking three indexes of the aggregation potential,the bacterial reduction ability,and the reduction degree(ID/IG)comprehensively.

Clinical significance of platelet mononuclear cell aggregates in patients with sepsis and acute respiratory distress syndrome

BACKGROUND The diagnosis of sepsis combined with acute respiratory distress syndrome(ARDS)has increased owing to the enhanced awareness among medical profes-sionals and the continuous development of modern medical technologies,while early diagnosis of ARDS still lacks specific biomarkers.One of the main patho-genic mechanisms of sepsis-associated ARDS involves the actions of various pathological injuries and inflammatory factors,such as platelet and white blood cells activation,leading to an increase of surface adhesion molecules.These adhesion molecules further form platelet-white blood cell aggregates,including platelet-mononuclear cell aggregates(PMAs).PMAs has been identified as one of the markers of platelet activation,here we hypothesize that PMAs might play a potential biomarker for the early diagnosis of this complication.METHODS We selected 72 hospitalized patients diagnosed with sepsis as the study population between March 2019 and March 2022.Among them,30 patients with sepsis and ARDS formed the study group,while 42 sepsis patients without ARDS comprised the control group.After diagnosis,venous blood samples were imme-diately collected from all patients.Flow cytometry was employed to analyze the expression of PMAs,platelet neutrophil aggregates(PNAs),and platelet aggregates(PLyAs)in the serum.Additionally,the Acute Physiology and Chronic Health Evaluation(APACHE)II score was calculated for each patient,and receiver operating characteristic curves were generated to assess diagnostic value.RESULTS The study found that the levels of PNAs and PLyAs in the serum of the study group were higher than those in the control group,but the difference was not statistically significant(P>0.05).However,the expression of PMAs in the serum of the study group was significantly upregulated(P<0.05)and positively correlated with the APACHE II score(r=0.671,P<0.05).When using PMAs as a diagnostic indicator,the area under the curve value was 0.957,indicating a high diagnostic value(P<0.05).Furthermore,the optimal cutoff value was 8.418%,with a diagnostic sensitivity of 0.819 and specificity of 0.947.CONCLUSION In summary,the serum levels of PMAs significantly increase in patients with sepsis and ARDS.Therefore,serum PMAs have the potential to become a new biomarker for clinically diagnosing sepsis complicated by ARDS.

Mechanical Behavior Based on Aggregates Microstructure of Ultra-high Performance Concrete

We developed ultra-high performance concrete(UHPC)incorporating mullite sand and brown corundum sand(BCS),and the quartz sand UHPC was utilized to prepare for comparison.The properties of compressive strength,elastic modulus,ultrasonic pulse velocity,flexural strength,and toughness were investigated.Scanning electron microscopy and nanoindentation were also conducted to reveal the underlying mechanisms affecting macroscopic performance.Due to the superior interface bonding properties between mullite sand and matrix,the compressive strength and flexural toughness of UHPC have been significantly improved.Mullite sand and BCS aggregates have higher stiffness than quartz sand,contributing to the excellent elastic modulus exhibited by UHPC.The stiffness and volume of aggregates have a more significant impact on the elastic modulus of UHPC than interface performance,and the latter contributes more to the strength of UHPC.This study will provide a reference for developing UHPC with superior elastic modulus for structural engineering.

Straw mulching alters the composition and loss of dissolved organic matter in farmland surface runoff by inhibiting the fragmentation of soil small macroaggregates

Straw mulching is a widespread practice for reducing the soil carbon loss caused by erosion.However,the effects of straw mulching on dissolved organic matter(DOM)runoff loss from black soil are not well studied.How straw mulching affects the composition and loss of runoff DOM by changing soil aggregates remains largely unclear.Here,a straw mulching treatment was compared to a no mulching treatment(as a control)on sloping farmland with black soil erosion in Northeast China.We divided the soil into large macroaggregates(>2 mm),small macroaggregates(0.25-2 mm),and microaggregates(<0.25 mm).After five rain events,the effects of straw mulching on the concentration(characterized by dissolved organic carbon(DoC)and composition(analyzed by fluorescence spectroscopy)of runoff and soil aggregate DOM were studied.The results showed that straw mulching reduced the runoff amount by 54.7%.Therefore,although straw mulching increased the average DOc concentration in runoff,it reduced the total runoff DOM loss by 48.3%.The composition of runoff DOM is similar to that of soil,as both contain humic-like acid and protein-like components.With straw mulching treatment,the protein-like components in small macroaggregates accumulated and the protein-like components in runoff declined with rain events.Fluorescence spectroscopy technology may help in understanding the hydrological paths of rain events by capturing the dynamic changes of runoff and soil DOM characteristics.A variation partitioning analysis(VPA)indicated that the DOM concentration and composition of microaggregates explained 68.2%of the change in runoff DOM from no mulching plots,while the change in runoff DOM from straw mulching plots was dominated by small macroaggregates at a rate of 55.1%.Taken together,our results demonstrated that straw mulching reduces the fragmentation of small macroaggregates and the loss of microaggregates,thus effecting DOM compositions in soil and reducing the DOM loss in runoff.These results provide a theoretical basis for reducing carbon loss in sloping farmland.

Online identification and extraction method of regional large-scale adjustable load-aggregation characteristics

This article introduces the concept of load aggregation,which involves a comprehensive analysis of loads to acquire their external characteristics for the purpose of modeling and analyzing power systems.The online identification method is a computer-involved approach for data collection,processing,and system identification,commonly used for adaptive control and prediction.This paper proposes a method for dynamically aggregating large-scale adjustable loads to support high proportions of new energy integration,aiming to study the aggregation characteristics of regional large-scale adjustable loads using online identification techniques and feature extraction methods.The experiment selected 300 central air conditioners as the research subject and analyzed their regulation characteristics,economic efficiency,and comfort.The experimental results show that as the adjustment time of the air conditioner increases from 5 minutes to 35 minutes,the stable adjustment quantity during the adjustment period decreases from 28.46 to 3.57,indicating that air conditioning loads can be controlled over a long period and have better adjustment effects in the short term.Overall,the experimental results of this paper demonstrate that analyzing the aggregation characteristics of regional large-scale adjustable loads using online identification techniques and feature extraction algorithms is effective.

Polycrystalline ZSM-5 Aggregates Induced by Seed and Catalytic Performance in Methanol to Hydrocarbon

Synthesis of ZSM-5 zeolite typically utilizes small molecule polyamines or quaternary ammonium salts as organic structure guiding agent(OSDA).By contrast,the OSDA-free hydrothermal synthesis system eliminates the use of organic templates and the subsequent calcination procedure.This not only reduces the cost of synthesis,but also prevents environmental pollution from the combustion of organic templates,representing an eco-friendly approach.Despite this,literature suggests that even so-called template-free synthesis systems often involve trace amount of organic substances like alcohol.In the present work,a calcined commercial ZSM-5 zeolite was served as seed,with sodium aluminate as aluminum source and silica sol as silicon source,ensuring an entirely template-free synthesis system.Polycrystalline ZSM-5 aggregates consisted of rod-like nanocrystals were successfully prepared in the completely OSDA-free system.Effects of the Si/Al ratio in ZSM-5 seed,dosage and crystallization conditions such as crystallization temperature and crystallization time on ZSM-5 synthesis were investigated.The results show that a highly crystallinity ZSM-5 aggregate consisting of primary nano-sized crystals less than 100 nm is produced from a gel precursor with 5.6%(in mass)seed after hydrothermal treatment for 48 h.Furthermore,the Si/Al ratio in ZSM-5 seed has little effect on the topological structure and pore structure of the synthesized samples.However,the seeds with a low Si/Al ratio facilitate faster crystallization of zeolite and enhance the acidity,especially the strong acid centers,of the catalyst.The catalytic performance of the synthesized polycrystalline ZSM-5 was evaluated during dehydration of methanol and compared with a commercial reference ZSM-5r.The results exhibit that as compared with the reference catalyst,the fabricated sample has a longer catalytic lifetime(16 h vs 8 h)attributed to its hierarchical pores derived from the loosely packed primary nanoparticles.Additionally,the prepared polycrystalline catalyst also exhibits a higher aromatics selectivity(28.1%-29.8%vs 26.5%).

Differentially Private Support Vector Machines with Knowledge Aggregation

With the widespread data collection and processing,privacy-preserving machine learning has become increasingly important in addressing privacy risks related to individuals.Support vector machine(SVM)is one of the most elementary learning models of machine learning.Privacy issues surrounding SVM classifier training have attracted increasing attention.In this paper,we investigate Differential Privacy-compliant Federated Machine Learning with Dimensionality Reduction,called FedDPDR-DPML,which greatly improves data utility while providing strong privacy guarantees.Considering in distributed learning scenarios,multiple participants usually hold unbalanced or small amounts of data.Therefore,FedDPDR-DPML enables multiple participants to collaboratively learn a global model based on weighted model averaging and knowledge aggregation and then the server distributes the global model to each participant to improve local data utility.Aiming at high-dimensional data,we adopt differential privacy in both the principal component analysis(PCA)-based dimensionality reduction phase and SVM classifiers training phase,which improves model accuracy while achieving strict differential privacy protection.Besides,we train Differential privacy(DP)-compliant SVM classifiers by adding noise to the objective function itself,thus leading to better data utility.Extensive experiments on three high-dimensional datasets demonstrate that FedDPDR-DPML can achieve high accuracy while ensuring strong privacy protection.

A non-probabilistic reliability topology optimization method based on aggregation function and matrix multiplication considering buckling response constraints

A non-probabilistic reliability topology optimization method is proposed based on the aggregation function and matrix multiplication.The expression of the geometric stiffness matrix is derived,the finite element linear buckling analysis is conducted,and the sensitivity solution of the linear buckling factor is achieved.For a specific problem in linear buckling topology optimization,a Heaviside projection function based on the exponential smooth growth is developed to eliminate the gray cells.The aggregation function method is used to consider the high-order eigenvalues,so as to obtain continuous sensitivity information and refined structural design.With cyclic matrix programming,a fast topology optimization method that can be used to efficiently obtain the unit assembly and sensitivity solution is conducted.To maximize the buckling load,under the constraint of the given buckling load,two types of topological optimization columns are constructed.The variable density method is used to achieve the topology optimization solution along with the moving asymptote optimization algorithm.The vertex method and the matching point method are used to carry out an uncertainty propagation analysis,and the non-probability reliability topology optimization method considering buckling responses is developed based on the transformation of non-probability reliability indices based on the characteristic distance.Finally,the differences in the structural topology optimization under different reliability degrees are illustrated by examples.

Effect of Modification Treatment on Chloride Ions Permeability and Microstructure of Recycled Brick-mixed Aggregate Concrete

The modification methods of pozzolan slurry combined with sodium silicate and silicon-based additive were respectively adopted to treat recycled coarse brick-mixed aggregate(RCBA)in this study.The compressive strength and chloride permeability resistance of recycled aggregate concrete(RAC)before and after modification treatment were tested,and the microstructure of RAC was analyzed by mercury intrusion porosimetry(MIP)and scanning electron microscopy(SEM).The results show that the physical properties of RCBA strengthened by modification treatment are improved,and the compressive strength and chloride permeability resistance of treated RAC are also significantly improved.The modification treatment optimizes the pore size distribution of RAC,which increases the number of gel pores and transition pores,and decreases the number of capillary pores and macro pores.The surface fractal dimension shows a significant correlation with chloride diffusion coefficient,indicating that the variation of chloride permeability of treated RAC is consistent with the microstructure evolution.

High-efficiency Carbonation Modification Methods of Recycled Coarse Aggregates

To solve the problem of only surface carbonation and realize high-efficiency carbonation of recycled coarse aggregate,the method of carbonated recycled coarse aggregate with nano materials pre-soaking was first put forward.The carbonation effect of modified recycled coarse aggregate with three different carbonation methods was evaluated,and water absorption,apparent density and crush index of modified recycled coarse aggregate were measured.Combined with XRD,SEM,and MIP microscopic analysis,the high-efficiency carbonation strengthening mechanism of modified recycled coarse aggregate was revealed.The experimental results show that,compared with the non-carbonated recycled coarse aggregate,the physical and microscopic properties of carbonated recycled coarse aggregate are improved.The method of carbonation with nano-SiO_(2) pre-soaking can realize the high-efficiency carbonation of recycled coarse aggregate,for modified recycled coarse aggregate with the method,water absorption is reduced by 23.03%,porosity is reduced by 44.06%,and the average pore diameter is 21.82 nm.The high-efficiency carbonation strengthening mechanism show that the pre-socked nano-SiO_(2) is bound to the hydration product Ca(OH)_(2) of the old mortar with nano-scale C-S-H,which can improve the CO_(2) absorption rate,accelerate the carbonation reaction,generate more stable CaCO_(3) and nano-scale silica gel,and bond to the dense three-dimensional network structure to realize the bidirectional enhancement of nano-materials and pressurized carbonation.It is concluded that the method of carbonation with nano-SiO_(2) pre-soaking is a novel high-efficiency carbonation modification of recycled coarse aggregate.

"Three-in-One" Ecological Development and Efficient Utilization Model of Aggregates

Green mining and the formation of an effective and efficient development model have become key issues that aggregates enterprises around the world need to solve urgently.On the basis of analyzing the development status of aggregates industry in Xiluodu area,the paper studied the main problems faced in the construction of green aggregates mines at present,and proposed a"three-in-one"ecological,intelligent and efficient green mine construction model for"ecological development","green logistics"and"solid waste recycling"of aggregates.The study has certain theoretical value and practical significance for the construction of green aggregates mine in Xiluodu area.

Repair of Second-Generation Recycled Fine Aggregate of Waste Concrete from Freeze-Thaw Environment by Carbonation Treatment

The reuse of waste recycled concrete from harsh environments has become a research hotspot in the field of construction.This study investigated the repair effect of carbonation treatment on second-generation recycled fine aggregate(SRFA)obtained from recycled fine aggregate concrete(RFAC)subjected to freeze-thaw(FT)cycles.Before and after carbonation,the properties of SRFA were evaluated.Carbonated second-generation recycled fine aggregate(CSRFA)at five substitution rates(0%,25%,50%,75%,100%)to replace SRFA was used to prepare carbonated second-generation recycled fine aggregate concrete(CSRFAC).The water absorption,porosity and mechanical properties of CSRFAC were tested,and its frost-resisting durability was evaluated.The results showed after carbonation treatment,the physical properties of SRFA was improved and met the requirements of II aggregate.The micro-hardness of the interfacial transition zone and attached mortar in CSRFA was 50.5%and 31.2%higher than that in SRFA,respectively.With the increase of CSRFA replacement rate,the water absorption and porosity of CSRFAC gradually decreased,and the mechanical properties and frost resistance of CSRFAC were gradually improved.Carbonation treatment effectively repairs the damage of SRFA caused by FT cycles and improves its application potential.

Cloud-Edge Collaborative Federated GAN Based Data Processing for IoT-Empowered Multi-Flow Integrated Energy Aggregation Dispatch

The convergence of Internet of Things(IoT),5G,and cloud collaboration offers tailored solutions to the rigorous demands of multi-flow integrated energy aggregation dispatch data processing.While generative adversarial networks(GANs)are instrumental in resource scheduling,their application in this domain is impeded by challenges such as convergence speed,inferior optimality searching capability,and the inability to learn from failed decision making feedbacks.Therefore,a cloud-edge collaborative federated GAN-based communication and computing resource scheduling algorithm with long-term constraint violation sensitiveness is proposed to address these challenges.The proposed algorithm facilitates real-time,energy-efficient data processing by optimizing transmission power control,data migration,and computing resource allocation.It employs federated learning for global parameter aggregation to enhance GAN parameter updating and dynamically adjusts GAN learning rates and global aggregation weights based on energy consumption constraint violations.Simulation results indicate that the proposed algorithm effectively reduces data processing latency,energy consumption,and convergence time.

Impact of wetting-drying cycles and acidic conditions on the soil aggregate stability of yellow‒brown soil

Soil aggregate is the basic structural unit of soil,which is the foundation for supporting ecosystem functions,while its composition and stability is significantly affected by the external environment.This study was conducted to explore the effect of external environment(wetting-drying cycles and acidic conditions)on the soil aggregate distribution and stability and identify the key soil physicochemical factors that affect the soil aggregate stability.The yellow‒brown soil from the Three Gorges Reservoir area(TGRA)was used,and 8 wetting-drying conditions(0,1,2,3,4,5,10 and 15 cycles)were simulated under 4 acidic conditions(pH=3,4,5 and 7).The particle size distribution and soil aggregate stability were determined by wet sieving method,the contribution of environmental factors(acid condition,wetting-drying cycle and their combined action)to the soil aggregate stability was clarified and the key soil physicochemical factors that affect the soil aggregate stability under wetting-drying cycles and acidic conditions were determined by using the Pearson’s correlation analysis,Partial least squares path modeling(PLS‒PM)and multiple linear regression analysis.The results indicate that wetting-drying cycles and acidic conditions have significant effects on the stability of soil aggregates,the soil aggregate stability gradually decreases with increasing number of wetting-drying cycles and it obviously decreases with the increase of acidity.Moreover,the combination of wetting-drying cycles and acidic conditions aggravate the reduction in the soil aggregate stability.The wetting-drying cycles,acidic conditions and their combined effect imposes significant impact on the soil aggregate stability,and the wetting-drying cycles exert the greatest influence.The soil aggregate stability is significantly correlated with the pH,Ca^(2+),Mg^(2+),maximum disintegration index(MDI)and soil bulk density(SBD).The PLS‒PM and multiple linear regression analysis further reveal that the soil aggregate stability is primarily influenced by SBD,Ca^(2+),and MDI.These results offer a scientific basis for understanding the soil aggregate breakdown mechanism and are helpful for clarifying the coupled effect of wetting-drying cycles and acid rain on terrestrial ecosystems in the TGRA.

Fake News Detection Based on Cross-Modal Message Aggregation and Gated Fusion Network

Social media has become increasingly significant in modern society,but it has also turned into a breeding ground for the propagation of misleading information,potentially causing a detrimental impact on public opinion and daily life.Compared to pure text content,multmodal content significantly increases the visibility and share ability of posts.This has made the search for efficient modality representations and cross-modal information interaction methods a key focus in the field of multimodal fake news detection.To effectively address the critical challenge of accurately detecting fake news on social media,this paper proposes a fake news detection model based on crossmodal message aggregation and a gated fusion network(MAGF).MAGF first uses BERT to extract cumulative textual feature representations and word-level features,applies Faster Region-based ConvolutionalNeuralNetwork(Faster R-CNN)to obtain image objects,and leverages ResNet-50 and Visual Geometry Group-19(VGG-19)to obtain image region features and global features.The image region features and word-level text features are then projected into a low-dimensional space to calculate a text-image affinity matrix for cross-modal message aggregation.The gated fusion network combines text and image region features to obtain adaptively aggregated features.The interaction matrix is derived through an attention mechanism and further integrated with global image features using a co-attention mechanism to producemultimodal representations.Finally,these fused features are fed into a classifier for news categorization.Experiments were conducted on two public datasets,Twitter and Weibo.Results show that the proposed model achieves accuracy rates of 91.8%and 88.7%on the two datasets,respectively,significantly outperforming traditional unimodal and existing multimodal models.