基于BiLSTM-Attention的F_(10.7)指数预测模型与中国自主数据集的应用

F_(10.7)指数是太阳活动的重要指标,准确预测F_(10.7)指数有助于预防和缓解太阳活动对无线电通信、导航和卫星通信等领域的影响.基于F_(10.7)射电流量的特性,在双向长短时记忆网络(Bidirectional Long Short-Term Memory Network,BiLSTM)基础上融入注意力机制(Attention),提出了一种基于BiLSTM-Attention的F_(10.7)预报模型.在加拿大DRAO数据集上其平均绝对误差(MAE)为5.38,平均绝对百分比误差(MAPE)控制在5%以内,相关系数(R)高达0.987,与其他RNN模型相比拥有优越的预测性能.针对中国廊坊L&S望远镜观测的F_(10.7)数据集,提出了一种转换平均校准(Conversion Average Calibration,CAC)方法进行数据预处理,处理后的数据与DRAO数据集具有较高的相关性.基于该数据集对比分析了RNN系列模型的预报效果,实验结果表明,BiLSTM-Attention和BiLSTM两种模型在预测F_(10.7)指数方面具有较好的优势,表现出较好的预测性能和稳定性.

An Assisted Diagnosis of Alzheimer’s Disease Incorporating Attention Mechanisms Med-3D Transfer Modeling

Alzheimer’s disease(AD)is a complex,progressive neurodegenerative disorder.The subtle and insidious onset of its pathogenesis makes early detection of a formidable challenge in both contemporary neuroscience and clinical practice.In this study,we introduce an advanced diagnostic methodology rooted in theMed-3D transfermodel and enhanced with an attention mechanism.We aim to improve the precision of AD diagnosis and facilitate its early identification.Initially,we employ a spatial normalization technique to address challenges like clarity degradation and unsaturation,which are commonly observed in imaging datasets.Subsequently,an attention mechanism is incorporated to selectively focus on the salient features within the imaging data.Building upon this foundation,we present the novelMed-3D transfermodel,designed to further elucidate and amplify the intricate features associated withADpathogenesis.Our proposedmodel has demonstrated promising results,achieving a classification accuracy of 92%.To emphasize the robustness and practicality of our approach,we introduce an adaptive‘hot-updating’auxiliary diagnostic system.This system not only enables continuous model training and optimization but also provides a dynamic platform to meet the real-time diagnostic and therapeutic demands of AD.

Pervasive Attentive Neural Network for Intelligent Image Classification Based on N-CDE’s

The utilization of visual attention enhances the performance of image classification tasks.Previous attentionbased models have demonstrated notable performance,but many of these models exhibit reduced accuracy when confronted with inter-class and intra-class similarities and differences.Neural-Controlled Differential Equations(N-CDE’s)and Neural Ordinary Differential Equations(NODE’s)are extensively utilized within this context.NCDE’s possesses the capacity to effectively illustrate both inter-class and intra-class similarities and differences with enhanced clarity.To this end,an attentive neural network has been proposed to generate attention maps,which uses two different types of N-CDE’s,one for adopting hidden layers and the other to generate attention values.Two distinct attention techniques are implemented including time-wise attention,also referred to as bottom N-CDE’s;and element-wise attention,called topN-CDE’s.Additionally,a trainingmethodology is proposed to guarantee that the training problem is sufficiently presented.Two classification tasks including fine-grained visual classification andmulti-label classification,are utilized to evaluate the proposedmodel.The proposedmethodology is employed on five publicly available datasets,including CUB-200-2011,ImageNet-1K,PASCAL VOC 2007,PASCAL VOC 2012,and MS COCO.The obtained visualizations have demonstrated that N-CDE’s are better appropriate for attention-based activities in comparison to conventional NODE’s.

Evolution and Governance of the Structure of Marine Economic Networks in China’s Coastal Areas: Based on Sea-related A-share Listed Companies

Data on discrete,isolated attributes of the marine economy are often used in traditional marine economic research.However,as the focus of urban research shifts from internal static attributes to external dynamic linkages,the importance of marine economic net-work research is beginning to emerge.The construction of the marine economic network in China’s coastal areas is necessary to change the flow of land and sea resources and optimize regional marine economic development.Employing data from headquarters and branches of sea-related A-share listed enterprises to construct the marine economic network in China,we use social network analysis(SNA)to discuss the characteristics of its evolution as of 2010,2015,and 2020 and its governance.The following results were obtained.1)In terms of topological characteristics,the scale of the marine economic network in China’s coastal areas has accelerated and expan-ded,and the connections have become increasingly close;thus,this development has complex network characteristics.2)In terms of spatial structure,the intensity of the connection fluctuates and does not form stable development support;the group structure gradually becomes clear,but the overall pattern is fragmented;there are spatial differences in marine economic agglomeration radiation;the radi-ation effect of the eastern marine economic circle is obvious;and the polarization effect of northern and southern marine economic circles is significant.On this basis,we construct a framework for the governance of a marine economic network with the market,the government,and industry as the three governing bodies.By clarifying the driving factors and building objectives of marine economic network construction,this study aims to foster the high-quality development of China’s marine economy.

RT-YOLO:A Residual Feature Fusion Triple Attention Network for Aerial Image Target Detection

In recent years,target detection of aerial images of unmannedaerial vehicle(UAV)has become one of the hottest topics.However,targetdetection of UAV aerial images often presents false detection and misseddetection.We proposed a modified you only look once(YOLO)model toimprove the problems arising in object detection in UAV aerial images:(1)A new residual structure is designed to improve the ability to extract featuresby enhancing the fusion of the inner features of the single layer.At the sametime,triplet attention module is added to strengthen the connection betweenspace and channel and better retain important feature information.(2)Thefeature information is enriched by improving the multi-scale feature pyramidstructure and strengthening the feature fusion at different scales.(3)A newloss function is created and the diagonal penalty term of the anchor frame isintroduced to improve the speed of training and the accuracy of reasoning.The proposed model is called residual feature fusion triple attention YOLO(RT-YOLO).Experiments showed that the mean average precision(mAP)ofRT-YOLO is increased from 57.2%to 60.8%on the vehicle detection in aerialimage(VEDAI)dataset,and the mAP is also increased by 1.7%on the remotesensing object detection(RSOD)dataset.The results show that theRT-YOLOoutperforms other mainstream models in UAV aerial image object detection.

An Efficient 3D CNN Framework with Attention Mechanisms for Alzheimer’s Disease Classification

Neurodegeneration is the gradual deterioration and eventual death of brain cells,leading to progressive loss of structure and function of neurons in the brain and nervous system.Neurodegenerative disorders,such as Alzheimer’s,Huntington’s,Parkinson’s,amyotrophic lateral sclerosis,multiple system atrophy,and multiple sclerosis,are characterized by progressive deterioration of brain function,resulting in symptoms such as memory impairment,movement difficulties,and cognitive decline.Early diagnosis of these conditions is crucial to slowing down cell degeneration and reducing the severity of the diseases.Magnetic resonance imaging(MRI)is widely used by neurologists for diagnosing brain abnormalities.The majority of the research in this field focuses on processing the 2D images extracted from the 3D MRI volumetric scans for disease diagnosis.This might result in losing the volumetric information obtained from the whole brain MRI.To address this problem,a novel 3D-CNN architecture with an attention mechanism is proposed to classify whole-brain MRI images for Alzheimer’s disease(AD)detection.The 3D-CNN model uses channel and spatial attention mechanisms to extract relevant features and improve accuracy in identifying brain dysfunctions by focusing on specific regions of the brain.The pipeline takes pre-processed MRI volumetric scans as input,and the 3D-CNN model leverages both channel and spatial attention mechanisms to extract precise feature representations of the input MRI volume for accurate classification.The present study utilizes the publicly available Alzheimer’s disease Neuroimaging Initiative(ADNI)dataset,which has three image classes:Mild Cognitive Impairment(MCI),Cognitive Normal(CN),and AD affected.The proposed approach achieves an overall accuracy of 79%when classifying three classes and an average accuracy of 87%when identifying AD and the other two classes.The findings reveal that 3D-CNN models with an attention mechanism exhibit significantly higher classification performance compared to other models,highlighting the potential of deep learning algorithms to aid in the early detection and prediction of AD.

基于改进YOLOv5s的输电线路螺栓缺销检测方法

针对无人机输电线路巡检图像中螺栓缺销检测精度较低、漏检较多的问题,提出了一种基于改进YOLOv5s的输电线路螺栓缺销检测方法。在Backbone部分嵌入Coordinate Attention注意力模块;在Neck部分原有的“FPN+PAN”结构的基础上,新增一条“自顶向下”的特征信息传递路径,跨越临近的同尺度特征层,与较浅层网络以加权融合的方式进行特征融合;将Head部分设置为解耦检测头,将对螺栓检测的分类任务与定位任务分开进行。改进后的YOLOv5s算法增强了对螺栓特征信息的学习能力。使用本方法在螺栓缺销数据集上实验,精确率提升了2.3%,召回率提升了3.4%,平均精度提升了3.1%,检测速度达到了41.1帧/秒,表明改进后的方法能提升输电线路螺栓缺销的检测能力,在智能巡检中具有一定的应用价值。

基于FasterNet和YOLOv5改进的玻璃绝缘子自爆缺陷快速检测方法

为了实现对电力输电线路中绝缘子缺陷实时快速的巡检需求,提出了一种结合FasterNet-tiny和YOLOv5-s-v6.1网络模型改进的缺陷快速检测算法FasterNet-YOLOv5。首先引入参数量小推理速度更快的FasterNet网络替换原先的CSPDarkNet53主干网络,加快网络的检测速度。然后结合由GhostNetv2网络提出的解耦全连接注意力机制(decoupled fully connected,DFC),在主干特征提取网络中设计了DFC-FasterNet模块,模块中的DFC Attention机制可以在特征提取过程中增大感受野,提升网络的检测精度。最后针对玻璃绝缘子自爆缺陷目标较小和背景较复杂的情况,重新设计Neck模块,提出BiFPN-F特征融合模块,使网络更精确地定位绝缘子缺陷区域。实验结果表明:改进后的算法可以快速精准定位,其均值平均精度(mean average precision,mAP)达到93.3%,相较于改进前提升5.67%,检测速度达到45.7 Hz,较改进前提升近1倍。同时与最新的YOLOv8n和YOLOv7-tiny相比,改进后的FasterNet-YOLOv5在自爆缺陷上的检测精度和速度更具优势,该文所提算法能够更快速地对绝缘子及其自爆缺陷实时定位识别。

ESG表现对制造业企业创新水平的影响机制研究

党的二十大报告指出坚持把发展经济的着力点放在实体经济上,而制造业是实体经济高水平发展的重要支撑。如何以创新驱动制造业高质量发展提升实体经济发展水平是我国当前发展的战略重点。同时,作为考察企业在环境、社会和治理方面表现的综合指标,环境、社会和治理(ESG)指标对企业的创新和绿色发展有重要影响。鉴于此,本研究以2009—2021年中国沪深A股上市公司中的制造业企业为研究样本,建立面板双向固定效应模型,从微观视角实证检验ESG表现对制造业企业创新水平的影响及作用机制。结果表明:良好的ESG表现可以有效提升制造业企业的创新水平,媒体关注度和政府补助在其中发挥部分中介作用。并且,这一促进作用在国有企业和大型企业中更为显著;东部地区企业的ESG表现对企业实质性创新水平的提升效果更强,而中西部地区企业的ESG表现对企业策略性创新的促进效应更强。为促进企业创新,助力实体经济实现高质量发展,应鼓励企业积极承担社会责任,健全ESG信息披露制度和表现评价体系,引导投资者树立ESG投资理念。

ESG评级分歧对企业绿色创新效率的影响研究

ESG评级作为指导企业战略决策的重要指标,评级结果的不一致对企业的投资决策会造成怎样的影响?本文选取2016~2022年沪深A股上市公司为研究样本,实证检验ESG评级分歧对企业绿色创新效率的影响。研究发现:ESG评级分歧越大,企业绿色创新效率越低;债务资本成本在ESG评级分歧与绿色创新效率之间发挥了部分中介作用;在分析师关注度和治理水平较高的企业中,ESG评级分歧对绿色创新效率的负向影响被削弱。本文将现有研究从ESG表现延伸至ESG评级分歧领域,为完善ESG评级体系建设、提高企业绿色创新效率提供了参考与启示。

改进YOLOX-s的密集垃圾检测方法

针对密集堆放的多种类垃圾检测存在识别率低、定位不够准确和待测目标被误检、漏检问题,提出了一种融合多头自注意力机制改进YOLOX-s的垃圾检测方法。在特征提取网络嵌入SwinTransformer模块,引入基于滑窗操作的多头自注意力机制,使得网络兼顾全局特征信息和重点特征信息,减少误检现象;在预测输出网络中使用可变形卷积,对初始预测框进行精细化处理,提高定位精度;在EIoU损失的基础上引入加权系数,提出加权IoU-EIoU损失,自适应调整训练时不同阶段不同损失的关注程度,进一步加快训练网络的收敛速度。在公开204类垃圾检测数据集中进行测试,结果表明,所提改进算法的平均精度均值分别可达80.5%和92.5%,优于当前流行目标检测算法,且检测速度快,满足实时性需求。

ESG信息披露、外部关注与企业风险

以中国A股2011~2020年上市公司为研究对象,实证检验了ESG信息披露对企业风险的影响效应与影响机制。研究发现:积极进行ESG信息披露能够降低企业风险,外部的媒体关注和分析师关注均强化了ESG信息披露的风险降低效应。机制检验表明:ESG信息披露通过增强利益相关者认同、减少信息不对称和管理ESG风险来降低企业风险水平。异质性检验还发现,ESG信息披露对企业风险的降低效应在非国有企业、成长期和成熟期企业样本中更显著。本研究有助于理解企业通过ESG战略进行风险管理的内在逻辑,为监管机构强化ESG信息披露指引以提升上市公司质量提供了经验证据。

Review of the evidence for the management of co-morbid Tic disorders in children and adolescents with attention deficit hyperactivity disorder

Attention deficit hyperactivity disorder(ADHD) is the most common neurodevelopmental disorder in children and adolescents, with prevalence ranging between 5% and 12% in the developed countries. Tic disorders(TD) are common co-morbidities in paediatric ADHD patients with or without pharmacotherapy treatment. There has been conflicting evidence of the role of psychostimulants in either precipitating or exacerbating TDs in ADHD patients. We carried out a literature review relating to the management of TDs in children and adolescents with ADHD through a comprehensive search of MEDLINE, EMBASE, CINAHL and Cochrane databases. No quantitative synthesis(meta-analysis) was deemed appropriate. Metaanalysis of controlled trials does not support an association between new onset or worsening of tics and normal doses of psychostimulant use. Supratherapeutic doses of dextroamphetamine have been shown to exacerbate TD. Most tics are mild or moderate and respond to psychoeducation and behavioural management. Level A evidence support the use of alpha adrenergic agonists, including Clonidine and Guanfacine, reuptake noradrenenaline inhibitors(Atomoxetine) and stimulants(Methylphenidate and Dexamphetamines) for the treatment of Tics and comorbid ADHD. Priority should be given to the management of co-morbid Tourette's syndrome(TS) or severely disabling tics in children and adolescents with ADHD. Severe TDs may require antipsychotic treatment. Antipsychotics, especially Aripiprazole, are safe and effective treatment for TS or severe Tics, but they only moderately control the co-occurring ADHD symptomatology. Short vignettes of different common clinical scenarios are presented to help clinicians determine the most appropriate treatment to consider in each patient presenting with ADHD and co-morbid TDs.

基于改进YOLOX-S的苹果成熟度检测方法

准确检测果园中未成熟与成熟的苹果对果园早期作物的负荷管理至关重要,提出一种能够实时检测苹果成熟度,并估算出整棵果树果实数量的方法。为提高YOLOX-S网络在复杂场景下的检测能力,在FPN(特征金字塔)的残差连接处增加了CoordinateAttention(位置注意力);为更好地检测图像中生长密集、存在遮挡、尺寸较小的苹果,将位置损失函数IoU_Loss更换为CIoU_Loss。试验结果表明,所提出的改进YOLOX-S检测算法相较于原算法,mAP值提高约1.97%,苹果低成熟度、中等成熟度和高等成熟度的AP值分别为90.85%、95.10%和80.50%。

改进YOLOv7-tiny与D-S理论结合的实验室人员行为检测研究

针对目前实验室场景缺少对人员行为检测的方法,且主流算法精度低、误检率高的问题,文中提出一种改进YOLOv7-tiny的人员行为检测算法,并通过多源信息融合,提高人员行为在实际实验室场景中的识别准确率。首先,在检测算法主干网络引入GhostNetV2轻量化网络,进一步降低模型计算量和复杂度;其次,在颈部网络嵌入改进后的CBAM_E注意力模块,加强目标重要特征的提取;再次,在预测端使用SIoU替换原有的损失函数,减少角度因素和边界框回归精度的影响。检测结果表明,相较于YOLOv7-tiny,文中算法精度提升10.08%,模型参数量和复杂度分别下降36.45%和46.76%。最后通过将检测数据与传感器采集数据运用D-S证据理论进行信息融合后发现,人员不规范行为检测的误检率得到有效降低。结果表明,该方法可实现对实验室人员不规范行为的有效检测。

The Impact of Government Supervision and Consumer Purchasing Behavior on Food Safety

This paper aims at the theoretical analysis to the impact of government supervision and consumer purchasing behavior on food quality security, so as to look for safety strategies and measures to strengthen and improve the level of food safety in China. Reputation mechanism is introduced and Bayesian approach is based on, in which government supervision as well as consumer purchasing behavior is taken as crucial factors to impact on the food quality security. As to the proposed quantitative indicators, government supervision includes exposure rate, fine and etc.;at the same time, consumer purchasing behavior includes consumer’s WTP for security food and consumer expectations to food safety. Taking China’s dairy industry as an example, it makes simulation by Netlog. The results show that consumer purchasing behavior alone has little effect on the dairy companies’ decision-making to be honest or counterfeiting enterprises. However, combination government supervision with purchasing behavior has great impact, and plays very good effects on food safety.

基于词嵌入和BiLCNN-Attention混合模型的政务文本分类方法

针对政务文本分析语境复杂、分类准确率低的问题,提出基于BERT词嵌入和BiLCNN-Attention混合模型的文本分类方法。首先采用BERT模型对政务文本进行词嵌入向量表示,然后混合使用双向长短时记忆网络BiLSTM和卷积神经网络CNN,同时引入注意力机制进行特征提取,融合了时序特征及局部特征并使特征得到强化,最后使用Softmax进行文本分类。实验表明,BERT词嵌入处理后混合模型的准确率较CNN和BiLSTM模型分别提升了3.9%和2.51%。

Government's Responsibility for Control of Agricultural Environmental Pollution in China

In this paper,the current situation of agricultural environmental pollution caused by the increase of chemicals input in agriculture in China was analyzed,and it is found that agricultural pollution is related to agricultural industrial policies,urban-rural economic structure,funds input in pollution control,comprehensive environmental management,laws of pollution control,and so forth. To control agricultural pollution effectively,it is needed to implement integration of agricultural and environmental policies,establish environmentally friendly agricultural technology popularizing system,implement integrated planning and management of a basin,and set up and improve legislation to protect agricultural environment.

ESG责任履行、外部治理与工业上市公司高质量发展

为探究ESG责任履行对工业上市公司高质量发展的影响,以及外部治理因素对二者关系的调节作用,以2011—2021年中国A股工业上市公司数据为样本,构建多维固定效应模型进行实证分析。研究发现,履行ESG责任有助于推动工业上市公司实现高质量发展,且公司治理责任的推动作用最大;分析师关注、媒体关注及环境规制等外部治理因素在ESG责任履行与高质量发展的关系中发挥正向调节作用。结果表明,工业上市公司应当加大ESG责任履行力度,可利用分析师关注、媒体关注激励工业上市公司履行ESG责任,政府及监管部门应当完善与ESG责任履行相关的法律法规,进而推动工业上市公司实现高质量发展。

China's Largest Granite-Type Gas Field was Discovered in Qinghai——The Inorganic Theory has Aroused Attention again

A growing number of oil and gas reservoirs have been discovered in granite and metamorphic crystallized rock areas. Statistics show that, about 157 oil and gas fields were found in crystallized bedrocks, with oil reserves of 5048x 10^8 t, and gas reserves of 2681x10^8m3. Among the discovered industrial oil and gas fields hosted in crystallized rocks, most occurred in granite rocks, occupying 40% in quantity and 75% in reserves, followed by those hosted in mafic and ultra-mafic rocks （about 3%）, and then tbllowed by those in volcanic rocks and metamorphic rocks.