关键词： deep-learning   convolution neural networks (CNNs)   artificial neural network   RapidEye   landslide mapping   mean intersection-over-union (mIOU)  

摘要： There is a growing demand for detailed and accurate landslide maps and inventories around the globe, but particularly in hazard-prone regions such as the Himalayas. Most standard mapping methods require expert knowledge, supervision and fieldwork. In this study, we use optical data from the Rapid Eye satellite and topographic factors to analyze the potential of machine learning methods, i.e., artificial neural network (ANN), support vector machines (SVM) and random forest (RF), and different deep-learning convolution neural networks (CNNs) for landslide detection. We use two training zones and one test zone to independently evaluate the performance of different methods in the highly landslide-prone Rasuwa district in Nepal. Twenty different maps are created using ANN, SVM and RF and different CNN instantiations and are compared against the results of extensive fieldwork through a mean intersection-over-union (mIOU) and other common metrics. This accuracy assessment yields the best result of 78.26% mIOU for a small window size CNN, which uses spectral information only. The additional information from a 5 m digital elevation model helps to discriminate between human settlements and landslides but does not improve the overall classification accuracy. CNNs do not automatically outperform ANN, SVM and RF, although this is sometimes claimed. Rather, the performance of CNNs strongly depends on their design, i.e., layer depth, input window sizes and training strategies. Here, we conclude that the CNN method is still in its infancy as most researchers will either use predefined parameters in solutions like Google TensorFlow or will apply different settings in a trial-and-error manner. Nevertheless, deep-learning can improve landslide mapping in the future if the effects of the different designs are better understood, enough training samples exist, and the effects of augmentation strategies to artificially increase the number of existing samples are better understood.

关键词： histotype classification   imaging mass spectrometry   machine learning   ovarian cancer  

摘要： Purpose Precise histological classification of epithelial ovarian cancer (EOC) has immanent diagnostic and therapeutic consequences, but remains challenging in histological routine. The aim of this pilot study is to examine the potential of matrix-assisted laser desorption/ionization (MALDI) imaging mass spectrometry in combination with machine learning methods to classify EOC histological subtypes from tissue microarray. Experimental design Formalin-fixed-paraffin-embedded tissue of 20 patients with ovarian clear-cell, 14 low-grade serous, 19 high-grade serous ovarian carcinomas, and 14 serous borderline tumors are analyzed using MALDI-Imaging. Classifications are computed by linear discriminant analysis (LDA), support vector machines with linear (SVM-lin) and radial basis function kernels (SVM-rbf), a neural network (NN), and a convolutional neural network (CNN). Results MALDI-Imaging and machine learning methods result in classification of EOC histotypes with mean accuracy of 80% for LDA, 80% SVM-lin, 74% SVM-rbf, 83% NN, and 85% CNN. Based on sensitivity (69-100%) and specificity (90-99%), CCN and NN are most suited to EOC classification. Conclusion and clinical relevance: The pilot study demonstrates the potential of MALDI-Imaging derived proteomic classifiers in combination with machine learning algorithms to discriminate EOC histotypes. Applications may support the development of new prognostic parameters in the assessment of EOC.

关键词： Discrete wavelet transform (DWT)   Electroencephalogram (EEG)   Epileptic seizure   Genetic algorithm (GA)   Particle swarm optimization (PSO)   Support vector machines (SVMs)  

摘要： The aim of this study is to establish a hybrid model for epileptic seizure detection with genetic algorithm (GA) and particle swarm optimization (PSO) to determine the optimum parameters of support vector machines (SVMs) for classification of EEG data. SVMs are one of the robust machine learning techniques and have been extensively used in many application areas. The kernel parameter's setting for SVMs in training process effects the classification accuracy. We used GA- and PSO-based approach to optimize the SVM parameters. Compared to the GA algorithm, the PSO-based approach significantly improves the classification accuracy. It is shown that the proposed Hybrid SVM can reach a classification accuracy of up to 99.38% for the EEG datasets. Hence, the proposed Hybrid SVM is an efficient tool for neuroscientists to detect epileptic seizure in EEG.

关键词： Counseling psychology   Computer science  

摘要： Motivational Interviewing (MI) is an evidence-based communication technique to increase intrinsic motivation and self-efficacy for behavior change. This goal is achieved through the exploration of the patient's own desires, ability, reasons, need for and commitment to the targeted behavior change. However, communication science approaches to understanding the efficacy of MI are inherently limited by traditional qualitative coding methods which is a time-consuming and resource-intensive process. Thus, an efficient method is required to automate the coding process which will accelerate the pace of communication research in behavioral science. The specific provider behaviors responsible for the elicitation of change talk, are also less clear and may vary by treatment context. Therefore, new design objective and perspective are necessary to understand which provider behaviors and in which contexts lead to patient change talk. In this dissertation, we deal with two types of clinical conversation, one that involves a face to face dialogue between patient and counselor and another one which involves an email-based conversation between patient and an ecoach. First, we leverage eight supervised machine learning models to automatically annotate counseling sessions with 37 African American adolescents with obesity and their caregivers. We examine the performance of classifiers using lexical, contextual, and semantic features, to predict the behavioral codes in the previously coded data. Second, understanding motivational interviewing mechanisms of effect, we focus on deep learning and probabilistic models and analyze the sequencing of patient-provider communication. The goal of these experiments is to identify the communication behaviors leading to the elicitation of client change talk, a marker of success in MI, and counter change talk, a marker of unsuccessful communication. Two approaches, recurrent neural networks and Markov models, were tested. As a continuation of our se

关键词： 量化投资   因子分析   机器学习方法   择时策略  

摘要： 决策树算法、随机森林和支持向量机模型是机器学习的细化分支领域,多年来深受各个学者、券商专业团队的喜爱,将其应用在量化投资领域。在此背景下,因子的构建和对于股票走势的判断,是这一领域专家学者不断追寻的话题。本文则希望通过研究和借鉴各个已有的文章的指标体系,结合相关机器学习算法,在此基础上进行创新和突破。本文把选股过程根据基本面指标和技术面指标的不同性质,将选股过程划分为两个阶段,以基本面指标判断上市公司长期状态和技术面指标判断买卖时机选择的理念进行本文投资组合的研究工作。首先,本文在第一章节阐述选题的背景与意义、相关文献综述、研究对象与研究工具;第二章节基于双因素模型的投资组合构造,围绕文中双因素模型投资组合策略构建思路进行展开,对本文提出的双因素模型进行具体介绍;第三章节内容围绕本文提出的双因素模型策略的第一个因素(阶段)即基本面因素进行实证研究;第四章节是对上节内容的优化和提升,在上节选出目标上市公司的基础上加入技术面因素(阶段),利用机器学习方法提升选股策略的效果;第五章节,对本文的研究内容结合具体实证情况进行总结。其次,本文将因子分析方法和决策树、随机森林、支持向量机方法分别应用在基本面选股和技术面选股(择时)两个阶段。结果显示:根据动态支持向量机模型实施后的交易策略能够获得较中证800指数3.29倍的期间累积收益和较全指消费指数3倍的期间累积收益,对应的期间累积收益率为106%。较之前仅由基本面选股,半年期间买入持有的等份额投资组合23.5%的累积收益率有大幅度的提升,前者的收益率相当于后者的4.51倍。除此之外,经过技术指标建模构造的投资组合在资金风险方面也较未经优化的模型有更好的表现。最后,本文基于2018年上半年度股票行情数据,从基本面角度在主要消费行业进行股票标的筛选,运用技术面指标对经由筛选后得出的股票标的涨跌情况进行逐日预测并根据预测结果进行相应的买卖操作,从而构建在期间的投资组合。对于完善选股投资策略的多样性有重要指导意义。

关键词： 多因子选股   随机森林   XGBOOST  

摘要： 量化投资是一种从历史数据提取出有效经验,并将这种经验应用于指导未来的一种投资方式。与传统的主观投资不同,量化投资不受人的主观情绪波动的影响,从而可以保持投资逻辑的稳定性和一致性。多因子选股作为权益类市场量化投资的重要策略一直以来都是学术界和业界关注的焦点。与传统的多因子选股模型通常关注标的资产的收益与各因子之间的线性关系不同,本文第一部分基于决策树模型可以发掘变量间的非线性和关系和解释性强的特点,对2006年4月至2018年6月的全A股市场数据建立了基于决策树的随机森林和XGBOOST两个集成算法模型的选股策略。并对上述策略分别进行了纯多头和市场中性的策略回测。其中18个月滚动训练的随机森林模型取得了纯多头23.8%的年化收益和0.673的夏普比率,市场中性26.7%的年化收益和1.447的夏普比率;18个月滚动训练的XGBOOST模型取得了纯多头26.8%的年化收益和0.782的夏普比率,市场中性28.9%的年化收益和1.594的夏普比率。纯多头策略和市场中性策略都远强于沪深300和中证500指数基准。第二部分首先提出了基于Focal-loss损失函数的XGBOOST模型以期在难分样本上取得更优效果,但结果并无明显变化。其次基于2017年市场风格切换导致的因子失效以及未来可能再次出现相同情况的现实,提出了基于不同长度训练集模型的模型融合方法,有效地提高了模型对于市场变化的反应能力。本文基于我国A股市场风格总体保持稳定的事实,放松了对于机器学习训练集和测试集样本来源同分布的假设。因此将时间序列数据截面化的思想的有效性仍有待市场样本外的表现加以验证。

关键词： 工业能源管理   工厂设备   能耗预测   机器学习   流式批处理  

摘要： 能源紧缺与环境污染等因素正在驱动制造行业探索更加高效的节能环保型生产。在第四次工业革命的背景下，将能耗预测功能用于工业能源管理的决策任务是改善工厂生产过程的有效途径。在赛博物理系统的层面上，可利用机器学习方法对广泛收集的制造信息数据进行分析，并将分析结果反馈给制造系统，作为一份决策报告用于生产过程优化。因此，基于机器学习方法的能耗预测系统具有重要的研究与应用意义。本文的研究目标旨在探索天气，设备工作时长和日期等因素对工厂设备能耗量的影响，并根据其历史数据建立一套可以用户自定义的模块化的能耗预测系统，使得系统在未来能够根据输入的当日天气等信息预测当日的设备能耗量，为能源管理提供前瞻性的决策支持。论文主要研究内容如下：　　1.基于CRISP-DM数据挖掘流程搭建一个针对设备能耗数据的数据处理流程，包括数据理解、数据预处理、建模和评估以及模型部署四个阶段。　　2.该系统实现了能耗信息的收集，在数据预处理和建模阶段设置可以用户自定义的模型及方法选择功能，使得用户可以在数据标准化，特征选择以及建模方法这三个步骤自行选择具体模型及方法，经选择后的模型在被导入Pipeline流式批处理管道后整合为一套机器学习模型。通过历史数据训练模型后，模型将自动生成各模块中的最优超参数。　　3.在用户可以自定义的步骤中对比各方法的性能表现，分析能耗的影响因素。包括特征选择阶段对缺失值比率、反向特征消除、主成分分析等方法的比较，以及在建模阶段对广义线性回归模型，决策树，支持向量机，多层感知机，决策树和集成学习方法的对比。　　4.系统在收集了2017年1月至2019年3月期间德国沃尔夫斯堡OpenHybrid实验室工厂所有设备每日的能耗数据，当地天气数据及日期信息后，将训练好的机器学习模型导入浏览器端的人机交互平台，用于未来对设备能耗的预测。

关键词： 电力巡检   计算机视觉   机器学习   小样本学习  

摘要： 巡检无人机成本低廉,智能化程度高,可对输电线路进行快速且安全的巡检工作,具有广阔应用前景,是当前电力巡检领域的热点研究项目。然而,电力巡检中类别繁杂,难以为每个类别都收集到足够的训练样本,且为训练样本人工标注的代价过于高昂。小样本分类算法适合于解决此类给定了丰富基础类别的训练样例,却缺乏新类标记实例的情况。但针对电力巡检这一具体的业务场景,仍缺乏相应研究。因此,需要进一步研究小样本分类算法,理清其原理、明确其数据基础,以期能将其成功应用于智能电力巡检系统。本文在经典分类神经网络模型的基础上进行了两个方法改进:增加了基于注意力机制的小样本分类权重发生器;将卷积神经网络模型的分类器更改为测试样本特征向量与类向量之间的余弦相似性函数。该方法在保留了其它小样本分类方法能较好地减少类内方差这一优点的基础上,对于较高域差异的训练集也体现了较好的自适应能力。综上所述,本文首先总结了近年来小样本学习方法的进展与成果,并阐述各算法的特点、主要针对的问题,比较分析了几种典型的小样本学习方法,包括匹配网络、原型网络、关系网络以及模型无关元学习方法。其次,设计了一个一致性的实验方案,减少因各小样本算法复杂的算法结构以及差异化的训练细节而导致的性能差异,并以此定量分析了各算法实际性能。此后探讨了小样本学习方法的训练目标,证明了更深的骨干网络显著降低了各方法之间的性能差异,并通过实验发现类内方差是造成小样本学习性能差异的主要因素之一。最后,实验结果还表明随着域差异增大,增强对于少数新类实例的自适应能力将比减少类内方差更能提高算法的准确率,该结果也对电力巡检领域数据集的制备工作提供了参考。

关键词： Alzheimer's disease   classification   early diagnosis   machine learning   magnetic resonance imaging   mild cognitive impairment   predictive analytics  

摘要： Background: Predicting clinical course of cognitive decline can boost clinical trials' power and improve our clinical decision-making. Machine learning (ML) algorithms are specifically designed for the purpose of prediction;however. identifying optimal features or algorithms is still a challenge. Objective: To investigate the accuracy of different ML methods and different features to classify cognitively normal (CN) individuals from Alzheimer's disease (AD) and to predict longitudinal outcome in participants with mild cognitive impairment (MCI). Methods: A total of 1,329 participants from the Alzheimer's Disease Neuroimaging Initiative (ADNI) were included: 424 CN, 656 MCI, and 249 AD individuals. Four feature-sets at baseline (hippocampal volume and volume of 47 cortical and subcortical regions with and without demographics and APOE4) and six machine learning methods (decision trees, support vector machines, K-nearest neighbor, ensemble linear discriminant, boosted trees, and random forests) were used to classify participants with normal cognition from participants with AD. Subsequently the model with best classification performance was used for predicting clinical outcome of MCI participants. Results: Ensemble linear discriminant models using demographics and all volumetric magnetic resonance imaging measures as feature-set showed the best performance in classification of CN versus AD participants (accuracy = 92.8%, sensitivity = 95.8%, and specificity = 88.3%). Prediction accuracy of future conversion from MCI to AD for this ensemble linear discriminant at 6, 12, 24, 36, and 48 months was 63.8% (sensitivity = 74.4, specificity = 63.1), 68.9% (sensitivity = 75.9, specificity = 67.8), 74.9% (sensitivity = 71.5, specificity = 76.3), 75.3%, (sensitivity = 65.2, specificity = 79.7), and 77.0% (sensitivity = 59.6, specificity = 86.1), respectively. Conclusions: Machine learning models trained for classification of CN versus AD can improve our prediction ability of MCI

关键词： 准确性   高效性   目标候选区域   行人检测  

摘要： 目标候选框可以完成对目标物体的初步定位,是目标检测和目标识别的基础,既适合对多类目标的定位也适合对单类目标的定位。在实时监控系统、门禁系统、刷脸支付等实际环境中具有重要作用。所以,目标候选框的准确高效定位具有重要研究意义。本文主要从目标候选框生成的高效性与准确性两个方面进行研究。(1)本文主要针对BING方法定位准确率不高的问题,首先使用扩展模板提取特征增强物体边缘特征的判别力,从而改进原梯度特征定位偏差的问题。然后采用高斯软权重非最大值抑制算法,解决了最大目标候选框附近非最大目标候选区域被强制归零的问题,提高了BING方法定位的准确率。最后在公开的数据集上对改进算法进行验证。实验结果表明,改进的BING算法在不增加算法计算复杂度的基础上,提高了BING方法定位的准确率。(2)本文基于具有较高准确率的选择性搜索方法进行两方面的改进,一是在初始分割阶段使用汉明距离计算相邻像素之间的相似性,二是针对选择性搜索方法合并策略计算复杂的问题,使用一种基于颜色的哈希策略的方法计算图像初始分割区域的相似度,从而判断相邻区域是否可以合并。这种方法大大地减少了选择性搜索算法在区域合并时计算的迭代次数,并在公开的测试集上进行验证。结果表明,本文改进的选择性搜索方法在尽可能不降低准确率的基础上,效率提高了60%。基于目标候选框在DPM行人检测算法中的应用。本文使用改进BING算法和改进选择性搜索算法代替DPM行人检测算法中的滑动窗口完成对行人的检测。通过行人数据集INRIA对检测结果进行验证,实验结果表明,使用改进目标候选框结合DPM的方法检测行人,加快了DPM行人检测的速度,实现了对行人的高效准确定位。