关键词： soil moisture   downscaling   SMAP   MODIS   machine learning  

摘要： Downscaling microwave remotely sensed soil moisture (SM) is an effective way to obtain spatial continuous SM with fine resolution for hydrological and agricultural applications on a regional scale. Downscaling factors and functions are two basic components of SM downscaling where the former is particularly important in the era of big data. Based on machine learning method, this study evaluated Land Surface Temperature (LST), Land surface Evaporative Efficiency (LEE), and geographical factors from Moderate Resolution Imaging Spectroradiometer (MODIS) products for downscaling SMAP (Soil Moisture Active and Passive) SM products. This study spans from 2015 to the end of 2018 and locates in the central United States. Original SMAP SM and in-situ SM at sparse networks and core validation sites were used as reference. Experiment results indicated that (1) LEE presented comparative performance with LST as downscaling factors;(2) adding geographical factors can significantly improve the performance of SM downscaling;(3) integrating LST, LEE, and geographical factors got the best performance;(4) using Z-score normalization or hyperbolic-tangent normalization methods did not change the above conclusions, neither did using support vector regression nor feed forward neural network methods. This study demonstrates the possibility of LEE as an alternative of LST for downscaling SM when there is no available LST due to cloud contamination. It also provides experimental evidence for adding geographical factors in the downscaling process.

关键词： Kernel logistic regression   Kernel functions   Hybrid models   Landslide   Geographic information system  

摘要： Globally, but especially in China, landslides are considered to be one of the most severe and significant natural hazards. In this study, bivariate statistical-based kernel logistic regression (KLR) models with different kernel functions (Polynomial, PUK, and Radial Basis Function), named the PLKLR, PUKLR, and RBFKLR models, were proposed for landslide susceptibility evaluation in Zichang City, China. Meanwhile, the present study aims to build landslide susceptibility maps based on bivariate statistical correlation analysis, optimization of different kernel functions, comparison of three landslide susceptibility maps and systematic analysis of spatial patterns. The steps of this article are organized as follows: Firstly, a landslide inventory containing 263 historical landslide locations was constructed. For the purpose of training and validation of models, 263 landslide locations were randomly divided into two parts with a ratio of 70/30. Secondly, 14 landslide conditioning factors were extracted from the spatial database. Subsequently, correlation analysis between the conditioning factors and the occurrence of landslides was conducted using frequency ratios. Then, the conditioning factors with normalized frequency ratios values were used as inputs to build the landslide susceptibility maps using the three models. A multicollinearity analysis was performed using collinearity statistics. Finally, the area under the receiver operating characteristic curve (AUC) was used for comparison and validation of models for recognizing the prediction capability. By further quantitative comparing mapped susceptibility values on a pixel-by-pixel basis, which can acquire underestimations and overestimations of factors (distance to river and slope) and susceptibility area. The results indicated that the PUKLR model had superior performance in landslide susceptibility assessment, with the highest AUC values of 0.884 and 0.766 for training and validation datasets, respectively. This

关键词： vegetation mapping   geo-objects   remote sensing   multi-source geospatial data   sample collection and purification  

摘要： Precise vegetation maps of mountainous areas are of great significance to grasp the situation of an ecological environment and forest resources. In this paper, while multi-source geospatial data can generally be quickly obtained at present, to realize effective vegetation mapping in mountainous areas when samples are difficult to collect due to their perilous terrain and inaccessible deep forest, we propose a novel and intelligent method of sample collection for machine-learning (ML)-based vegetation mapping. First, we employ geo-objects (i.e., polygons) from topographic partitioning and constrained segmentation as basic mapping units and formalize the problem as a supervised classification process using ML algorithms. Second, a previously available vegetation map with rough-scale label information is overlaid on the geo-object-level polygons, and candidate geo-object-based samples can be identified when all the grids' labels of vegetation types within the geo-objects are the same. Third, various kinds of geo-object-level features are extracted according to high-spatial-resolution remote sensing (HSR-RS) images and multi-source geospatial data. Some unreliable geo-object-based samples are rejected in the candidate set by comparing their features and the rules based on local expert knowledge. Finally, based on these automatically collected samples, we train the model using a random forest (RF)-based algorithm and classify all the geo-objects with labels of vegetation types. A case experiment of Taibai Mountain in China shows that the methodology has the ability to achieve good vegetation mapping results with the rapid and convenient sample collection scheme. The map with a finer geographic distribution pattern of vegetation could clearly promote the vegetation resources investigation and monitoring of the study area;thus, the methodological framework is worth popularizing in the mapping areas such as mountainous regions where the field survey sampling is difficult to

关键词： Machine learning   Whale optimization algorithm   Wavelet transform   Flood frequency   Climate change   Fuzzy uncertainty  

摘要： In the present study, for the first time, a new framework is used by combining metaheuristic algorithms, decomposition and machine learning for flood frequency analysis under climate-change conditions and application of HadCM3 (A2 and B2 scenarios), CGCM3 (A2 and A1B scenarios) and CanESM2 (RCP2.6, RCP4.5 and RCP8.5 scenarios) in global climate models (GCM). In the proposed framework, Multivariate Adaptive Regression Splines (MARS) and M5 Model tree are used for classification of precipitation (wet and dry days), whale optimization algorithm (WOA) is considered for training least square support vector machine (LSSVM), wavelet transform (WT) is used for decomposition of precipitation and temperature, LSSVM-WOA, LSSVM, K nearest neighbor (KNN) and artificial neural network (ANN) are performed for downscaling precipitation and temperature, and discharge is simulated under present period (1972-2000), near future (2020-2040) and far future (2070-2100). Log normal distribution is used for flood frequency analysis. Furthermore, analysis of variance (ANOVA) and fuzzy method are employed for uncertainty analysis. Karun3 Basin, in southwest of Iran, is considered as a case study. Results indicated that MARS performed better than M5 model tree. In downscaling, ANN and LSSVM_WOA slightly outperformed other machine learning algorithms. Results of simulating the discharge showed superiority of LSSVM_WOA_WT algorithm (Nash-Sutcliffe efficiency (NSE) = 0.911). Results of flood frequency analysis revealed that 200-year discharge decreases for all scenarios, except CanESM2 RCP2.6 scenario, in the near future. In the near and far future periods, it is obvious from ANOVA uncertainty analysis that hydrological models are one of the most important sources of uncertainty. Based on the fuzzy uncertainty analysis, HadCM3 model has lower uncertainty in higher return periods (up to 60% lower than other models in 1000-year return period).

关键词： Fracture toughness   Knitting fabric   Pure and hybrid composite   Finite element analysis   Machine learning algorithms  

摘要： In recent years, it became possible to use different methods for the analysis of mechanical systems with the help of computers to learn like humans and by increasing their interaction with the world by observing autonomously. One of these mechanical analyzes is the fracture mechanics in which the behavior of the laminated composites having a crack is examined. In this study, experimental methods, finite element analysis (FEA) and machine learning algorithms (MLA) were used to analyze the fracture behavior of polymer composites in Mode I, Mode I/II and Mode II loading situations. For the experimental study, the fracture behaviors of the laminated composites reinforced with pure glass, pure carbon and glass/carbon hybrid knitted fabrics were tested with the help of Arcan test apparatus. In the finite element method, the linear elastic fracture behavior at the crack tip was analyzed by using the J-integral method. In the field of MLA, there is no single learning algorithm that provides good learning on all real-world problem data. Therefore, algorithm selection is done experimentally so various machine algorithms were used in the study. The analysis result showed that the finite element analysis and machine learning results were in good agreement with experimental measurements. This study is particularly important for the comparison of machine learning techniques with FEA in regression applications.

关键词： 电磁泄漏   旁路攻击   机器学习   深度学习   迁移学习  

摘要： 电磁泄漏信号是电子系统正常运行期间,由内部电流的变化所产生的电磁辐射信号。在正常情况下,电阻、电源、计算芯片等电子元件都会产生电磁辐射并引起电磁泄漏。本文针对嵌入式设备产生的电磁泄漏信号进行分类识别,对于嵌入式设备的安全性评估有重要意义。本文主要有如下四个创新点:第一,搭建了基于单片机存储、SIMECK32/64进行加密过程的电磁泄漏信号自动采集平台和基于Jetson Nano的神经网络电磁泄漏信号的自动采集平台。并设计了对应的硬件组件和软件,使采集过程和数据筛选过程更加高效。第二,为了解决直接训练卷积神经网络梯度不下降的问题,结合汉明重模型理论,提出了一种新的卷积神经网络结构——简单任务网络(Easy Task Network),并在此网络结构的基础上,利用采集的单片机存储过程中的电磁泄漏,在存储任务数据集上进行验证并达到了96.80%的准确率。第三,在不知道任何明文、密文及泄漏中间值信息的情况下使用全随机密钥,并用SIMECK32/64算法进行加密,进行电磁泄漏攻击。使用设计的神经网络结构并利用基于标签的迁移学习,达到了 96%的平均准确率。并且通过不同任务的电磁泄漏信号的数据集进行迁移学习,目标数据集提高了 6%的分类识别准确率,证明了不同任务的电磁泄漏信号之间的可迁移性。第四,针对Jetson设备神经网络结构进行基于电磁泄漏的旁路攻击。进行了两个维度的安全评估,一种是对同一框架TensorRT下7种不同的神经网络的泄漏信号进行分类;另一种是对TensorRT、Tensorflow、Pytorch三种框架下同一神经网络的电磁泄漏进行分类。本任务中通过不同策略设计并测试了如何将二维卷积神经网络转换为一维卷积神经网络的结构,通过实验验证了其可行性,证明了良好的网络转换策略能够提升5%～12%的分类识别准确率。

关键词： Deep learning   Algorithms   Optimization   Artificial intelligence   Computer science  

摘要： The numerical reservoir simulation model is a stable among the tools used to address the problem of describing the fluid flow behavior of producing reservoir from wells. The downside of using these models is the inherent cost of development and deployment, especially for small fields, and resources restrained projects. The recent advances in machine learning methods in well testing domain, ignited an interest to bring these capability to reservoir simulation and management. In this work, supervised and unsupervised machine learning techniques were applied to reservoir management tasks, that are usually dealt with using a numerical reservoir simulation model. Starting with well level analysis, simulation of downhole pressure response in producing wells, as a function of production and /or injection rate, using only field data is investigated. The methods used include algorithms composed of feed forward, recurrent, and convolution layers. The same is used to simulate water cut response as a function of operational parameters. The results suggest that it is possible to generate accurate prediction using these techniques. Extending the analysis to field level, we showed how using unsupervised learning techniques helped in guiding samples selection for training, then we applied generative modeling techniques using variational autoencoder to the problem of spatial control in the reservoir form data. We compared the performance of it to autoencoder, and other machine learning algorithms to predict multiphase production profiles form wells. Our investigation indicates that it can be done successfully in undrilled locations. We further applied conditional variational autoencoder along with deep feature interpolation methods to generate novel simulations that are not available in the training data set, extending the range of available simulated profiles without additional training or the need to conduct numerical simulation runs. Finally, we showed, using real field examples,

关键词： 职业分类   劳动力市场大数据   职业标准   机器学习技术  

摘要： 劳动力市场在多种因素的作用下催化了新职业的产生和分化,就业形态也趋于多元化发展,需要公共部门做好职业管理。为此,我国逐步建立和健全了我国的职业分类体系,以满足经济社会发展和人力资源管理的需要。但是,在职业标准的描述上,还存在着内容不详、更新慢、指导性差等缺陷;劳动力市场职业分类标准不统一、职业名称不规范等问题也阻碍了职业分类管理在劳动力市场中的有效执行,以至于公共部门无法监测不同职业在数量和结构上的发展和变化,在决策过程中依然缺少可靠的抓手。近几年,以招聘广告为代表的劳动力市场大数据的出现拓展了数据类型和数据来源,既能够弥补传统统计调查方式结果数据的不足,也较为客观地反映了劳动力市场的整体情况,为改善职业分类管理创造了条件。 本文以比较有代表性的直播销售员为例,探讨了职业标准的构建过程,并在职业标准的基础之上,利用劳动力市场大数据和机器学习技术探索具有普适性的职业分类预测的实现路径。研究总共分为两个部分,第一阶段研究是构建直播销售员的职业标准,以劳动分工理论为基础的工作内容描述已经无法满足对某个职业进行客观详尽描述,而基于人职匹配理论将技能水平描述也纳入到职业标准的维度则丰富了职业标准的内容,通过对25名直播销售员从业者或者职业相关人员进行了深度访谈,并对访谈结果进行了整理、归纳、编码和演绎,逐步确定了直播销售员的职业标准。第二阶段研究是在职业标准基础上,利用劳动力市场大数据对直播销售员进行职业分类预测。通过爬虫软件收集到了2535条直播销售员的招聘广告,并运用机器学习技术中朴素贝叶斯文本分类构建了职业分类预测模型,对招聘广告是否真正属于直播销售员这一职业类别进行预测,并对分类预测效果进行了评估,结果表明该职业分类预测模型具有较高的准确性和稳定性,如果推广至所有职业,就能够基于现行职业分类体系对劳动力市场实现职业分类。 研究成果丰富了现有的职业分类管理领域的研究,深化了人职匹配理论的研究体系,不仅在内容上拓展了职业标准的维度,提升了职业标准的指导价值,也使得职业分类管理从空有标准转向有效落实和执行。

关键词： 机器学习方法   航空旅客出行指数   相关性理论   互信息模糊理论   贝叶斯网络   极值理论  

摘要： 近年来,机器学习与大数据技术取得了长足进步,实现了以机器学习方法为主并融合其他相关数据分析技术在各个领域的广泛应用。尤其在民航运输领域迅猛发展的背景下,由于智慧机场建设和航空旅客绿色出行的内在驱动,使得行业生产相关数据不断增长与积累,运用机器学习方法赋能机场运营管理与相关服务质量水平提升。然而,由于航空旅客出行相关问题的复杂性,关联的业务错综交织,采用传统的一些方法往往不能得到理想的结果。本文主要研究以机器学习方法为关键技术实现路线,构建管理与决策新模型方法,为航空旅客出行相关问题的解决提供更优方案。围绕航空旅客出行中的旅客吞吐量、航班旅客量、出行指数分析预测及地服保障人员排班等关键问题展开探讨,本文重点研究利用现有机器学习技术方法与相关数据分析技术,结合问题特点构造新模型方法,进而实现将这些新模型方法应用于解决机场运营管理与航空旅客出行相关具体问题。首先对智慧机场、航空旅客出行、旅客吞吐量、航线航班客座率、地服人员排班等问题进行梳理分析,同时对机器学习方法、模糊理论、信息论、极值理论等关键技术进行了探讨;然后基于时间序列、互信息、模糊、极值理论相互融合并改进机器学习方法模型,提出基于长短记忆与支持向量回归的模型方法、基于互信息改进支持向量回归模型方法、基于模糊最大化互信息改进贝叶斯网络模型方法;最后将这些方法分别应用于吞吐量预测、航班旅客量预测、航空旅客出行指数预测以及地服保障人员优化排班等难题,并取得了较好的效果。本文主要的研究工作说明如下:1)针对航空旅客吞吐量预测问题,根据航空旅客吞吐量预测时间序列分布特点,考虑长短期记忆网络(LSTM)改善了RNN中存在的长期依赖问题,提出了一种基于长短期记忆网络改进的支持向量回归预测方法即LSTM-SVR模型方法。凭借LSTM作为动态的循环神经网络,能比较好地反映航空旅客吞吐量动态变化结果,而SVR具有的修正特点即“消去”大量冗余样本,经过两种机器学习方法的优势融合,极大减少了预测误差随时间的传递。通过实验分析显示,LSTM-SVR模型方法可行且预测精度更高,该方法对机场规模与资源配套建设具有辅助决策意义。此外,根据机场网络布局,考虑机场网络航线吞吐量影响,构建航班旅客预测模型,为航空公司经营的中航班旅客量预测提供技术手段即可以有效帮助航空公司更了解客户,这一研究成果对航空公司航线的运营收益提升带来一定的帮助。2)针对航空旅客出行评价问题的研究,首先,本文首次提出航空旅客出行指数与指数等级概念,它们能很好地反映机场繁忙程度,为机场运营管理与旅客出行提供了一个新的参考指标。其次,针对航空旅客出行指数预测,利用互信息值选取关键影响因子,构建基于互信息改进的支持向量回归MI-SVR模型方法。最后,实证分析发现,通过互信息选择特征影响因子后,常见的预测方法预测效果都有明显提升,而本文提出的MI-SVR模型方法相对更优。该创新研究成果较好反映了航空旅客出行指数的时空分布,对机场运营管理意义重大,一方面,机场可以选择在相对不那么繁忙的时段内安排相关设施设备的运维或其他施工建设活动,最大限度减少此类工作给机场运营管理带来的影响;另一方面,针对旅客出行高峰时段,充分调配相应的内外部资源保障服务质量。航空旅客出行指数不仅适用小型机场,更适用于大型枢纽机场,对旅客选择民航方式出行时段具有参考价值。3)考虑影响出行指数因素中小概率情况,利用模糊思想提出了基于模糊最大化互信息思想改进的贝叶斯网络航空旅客出行指数预测方法。该方法通过模糊构建虚拟样本集即随机将数据进行分组,将偶然出现的现象数据“均衡化”处理洒落在混乱数据集中,利用数据原有的冗余数据减小,强化相关性数据的作用,达到提高分类器的分类效果。实证分析发现本文提出的改进贝叶斯网络模型方法可行且有效,该方法不仅适用具有大量航空旅客出行指数历史数据的精准预测,也适用少量数据时的精确预测,且表现更突出。此外,利用手肘法改进聚类方法K-means对出行指数、影响因素聚类分析,实证分析发现聚类效果良好。4)针对航空旅客出行指数极值分布下的地服保障人员排班问题,首先根据地服保障人员排班特点,结合极值理论与经典智能优化算法,提出了考虑航空旅客出行极值分布的地服保障人员排班模型;然后利用智能优化算法进行求解获得较优的地服保障人员排班方案。最后进行实证分析,实验结果表明,通过机器学习方法对出行指数预测获得极值分布后,运用智能优化算法排班比人工标准排班效率更高,且对比分析了是否考虑极值分布的情况下进行地服保障人员智能优化算法排班,考虑极值与按均值条件下排班,前者地服保障工作总量在既定的工作期内减少,人员利用效率得到提高。探讨了某机场地服保障服务53周为工作期,工作量(以小时计算)相比减少约8.1%,具有良好的运营效益。此外,通过问卷调查发现,科学的地服保障人员排班方案有利

关键词： 心肌淀粉样变   肥厚型心肌病   机器学习  

摘要： 目的:本研究提出基于机器学习方法结合二维斑点追踪超声心动图(2D-STE)数据进行心肌淀粉样变和肥厚型心肌病的智能鉴别研究。方法:选取2015年6月-2018年11月在浙江大学医学院附属第一医院住院治疗,经活检明确诊断CA的患者74例,选择同期入住的HCM患者64例作为对照。收集患者的临床资料,实验室检查,常规超声心动图和2D-STE等数据。利用机器学习算法分别建立2种模型:1、利用临床资料,实验室检查,常规超声心动图和斑点追踪超声数据通过支持向量机,随机森林和梯度提升等机器学习方法进行数据分析和构建预测模型,并与传统的Logistic回归模型相对比;2、利用2D-STE应变数据通过分布式梯度提升(light GBM),极端梯度提升(XGBoost)和整合两种算法进行数据分析和构建预测模型。模型建立后利用受试者工作特征曲线(ROC)及曲线下面积(AUC)评估不同模型的性能。结果:在利用临床资料,实验室检查,常规超声心动图和斑点追踪超声数据进行建模的模型中,机器学习方法均表现出良好的区分性能(AUC值均>0.9),其中极端梯度提升模型具有最高AUC值,为0.98。与传统的Logistic回归模型(AUC=0.91)相比,支持向量机(AUC=0.95,P=0.477),随机森林(AUC值=0.97,P=0.301)和梯度提升(AUC值0.98,P=0.230)等机器学习模型预测CA的性能呈现不同程度的改善趋势,但AUC值的差异没有统计学意义。在利用2D-STE应变数据进行建模的模型中,整合两种算法的投票模型的判别能力与分布式梯度提升相仿(两者AUC均为0.86),而极端梯度提升的AUC值稍低,为0.84。随后在进行5折交叉验证时,结果提示投票算法在全局上更加稳健,并在部分测试集上优于单个算法。在第1种模型中,偏心指数,左室射血分数(LVEF),相对心尖纵向应变(RELAPS),慢性肾病,室间隔厚度(LVSd)和收缩压是随机森林和梯度提升模型特征选择中前6位的最重要的预测因素。在第2种模型中,Rad Strain3(中间段室间隔径向应变)是最重要的预测因子,其次是Long Strain Epi7(心外膜基底段前壁纵向应变),Cir Strain R4(基底段后壁环向应变率)。结论:数据驱动的机器学习方法在鉴别心肌淀粉样变性和肥厚型心肌病方面具有良好的表现,可以在不作预先假设的情况下,自动整合大量的变量数据进行特征选择,从而识别出最具有鉴别意义的预测因子。在大数据背景下,自动化的机器学习方法将有助于更好的识别出心肌淀粉样变的患者,以便及时进行有效的干预,从而改善预后。