关键词： 中文微博   情感词典   情感分析   多特征融合   类序列规则   主题模型  

摘要： 在当今互联网高速发展的时代下,社交网络和移动终端技术的快速发展,兴起了很多社会媒体和社交网站平台,而微博就是其中最为流行的一种社交网络平台。微博作为一种社交网站的主要媒体形式,它短小、精悍、快捷,越来越受到人们的喜爱。人们也越来越倾向于通过微博获取新闻、评论、娱乐、知识学习等信息,不知不觉间,微博对网络舆情的传播施加了越来越重要的影响。由于用户热衷于通过微博来表达观点、态度和情感等,所以挖掘这些微博信息的情感对于政府的舆情监控、商家的市场分析、用户评论的分析与决策等等有着十分重要的意义。传统的微博情感分析即对短文本的情感分析,仅仅是将微博文本进行倾向性判断,最终只分析出微博的情感倾向性是褒义、中性还是贬义。由于微博短文本语言的特殊性,它与一般普通的文本不同,常常会出现极其简短、口语化、语法不规范、网络新词频繁出现等现象。因此,针对微博语言的这些现象,对微博进行更细粒度的研究显得尤为重要。本文将微博情感分为高兴、喜欢、生气、厌恶、恐惧、悲伤和惊讶七大类,分别使用基于词典与规则结合的语义加权方法和基于有监督的机器学习方法对选取的微博文本进行细粒度的情感分类,对微博文本的情感分析工作分为主题相关的情感分析和主题无关的情感分析。本文的主要研究内容如下：(1)对微博文本中的语言特点现象进行了研究,针对这些现象分析微博文本中各种对后面分类有利的特征,并介绍了目前情感分析的研究基础和现状,有关情感分析的国内外评测会议和情感分析的应用研究。(2)设计构建了文本的中文微博情感词典,情感词典的类别包括高兴、喜欢、生气、厌恶、恐惧、悲伤和惊讶七大类；除此之外,还构建了网络新词词典、程度副词词典、连词词典和表情符号的扩展。情感词典是以大连理工大学情感词汇本体为基础,对其进行扩充,并进行了基于大连理工大学的情感词典和文本词典的实验对比,实验结果证明了本论文构建的词典效果更好。(3)以构建的词典为基础,文本使用了基于词典与规则结合的语义加权算法对微博进行情感值的加权,以此作为基于词典与规则的方法,并与后面基于机器学习的方法进行对比。(4)本文提出一种基于类序列规则的多特征融合的情感分析方法。利用基于词典和传统SVM两种方法得到微博各个句子的两个标签,将微博文本表示成为序列形式,结合连词挖掘其中的序列规则作为特征再对分类器进行训练,最终利用提取的情感词特征、标点符号特征、句子结构特征和类序列规则特征进行训练,经过参数调节后得到分类效果较好的分类器。(5)针对主题相关的微博数据,还考虑到提取主题特征对训练的影响,实验证明,通过主题模型得到的概率分布作为特征也能提升最终的分类效果。(6)本文采用的数据集是COAE(中文倾向性分析评测)2014会议上提供的微博数据。训练数据包含4000条微博和14526个句子,测试集包含5000条微博和16785个句子,每条微博的情感类型均已被标注,所有的这些微博文本均来自新浪微博。最后进行了各种方法的实验对比。

关键词： 智能诊断   机器学习   疾病诊断模型   支持向量机   优化方法  

摘要： 研究疾病智能诊断模型，对提高疾病诊断的正确性、实时性，减轻医务人员的劳动强度，具有十分重要的意义。机器学习方法不仅可以对医疗检测出的数值结果进行主动学习，而且能在繁杂模型中快速的做出智能决策。近年来，机器学习方法被广泛应用于各种医疗诊断系统。然而，并非单一的机器学习方法在对未知数据进行预测时就能达到较高的泛化能力，在机器学习算法应用中，还需要综合考虑多种技术的融合和优化。　　论文在研究机器学习和疾病诊断方法的基础上，重点研究了以支持向量机(SVM)为基础的疾病诊断模型及其优化方法，主要研究内容和工作成果如下：　　1、对传统的机器学习诊断系统进行了深入分析，考虑到疾病数据具有不平衡、冗余和高维特性，研究了数据预处理，降维及不平衡数据处理方法，以典型分类方法为基础，分别建立了基于决策树与SVM的乳腺癌诊断模型，试验结果表明，SVM诊断方法表现了更好的预测性能。　　2、针对SVM的性能易受其超参数(如惩罚因子C和RBF核的核宽γ)影响，将网格参数寻优算法引入到SVM的诊断模型。考虑到网格参数寻优的低效性和敏感性，利用智能计算技术，分别设计了基于遗传算法(GA)和粒子群优化算法(PSO)的SVM诊断模型。实验结果表明该类模型可以有效地筛选出最重要的致病特征，准确地对乳腺癌病情做出诊断。　　3、集成学习因其能显著提高一个学习系统的泛化能力而得到了机器学习界的广泛关注。利用Adaboost方法改变样本分布，获得一组具有差异性的基分类器，从而建立基于Adaboost的SVM集成诊断模型，并设计了实验验证。另外，在分析集成学习及选择性集成方法缺陷的基础上，尝试从准确性与多样性二者均衡的角度，对集成系统质量评价问题展开研究，提出了一种新的评价策略FDA（F-Measure based on Diversity and Accuracy，FDA），该策略源于信息检索F值均衡思想，为有效提高集成系统泛化能力提供了参考。　　本文所提出的三种智能疾病诊断模型，都分别在UCI数据资源库中的疾病数据集上得到了验证。

摘要： Machine learning and data mining are currently hot topics of research and are applied in database, artificial intelligence, statistics, and so on to discover valuable knowledge and the patterns in big data available to users. Data mining is predominantly about processing unstructured data and extracting meaningful information from them for end users to help take business decisions. Machine learning techniques use mathematical algorithms to find a pattern or extract meaning out from big data. The popularity of such techniques in analyzing business problems has been enhanced by the arrival of big data. The main objective of this thesis is to study the importance of big data and machine learning and their impact on transportation industry. This thesis is primarily a review of the important machine learning algorithms and their applications in the field of big data. The author has tried to showcase the need to extract meaningful information from the vast amount of big data in the form of traffic data available in today’s world and also listed different machine learning techniques that can be used to extract this knowledge required in order to facilitate better decision making for transportation applications. The analysis is done by using five different multivariate analysis and machine learning techniques in data mining namely cluster analysis, multivariate linear regression, hierarchical multiple regression, factor analysis and discriminant analysis in two different software packages namely SPSS and R. As part of the analysis, the author has tried to explain how knowledge extracted from random traffic data containing variables such as age of the driver, sex of the driver, the day of the week, atmospheric condition and blood alcohol content of the driver can play an important role in predicting the traffic crash. The data taken into account is accident data, which was obtained from Fatality Analysis Reporting System (FARS) ranging from the year 1999 to 2009. It is concl

关键词： MACHINE learning -- Research   SUPPORT vector machines   RESEARCH   DECISION making   ARTIFICIAL intelligence   LAND cover  

摘要： Until now, few research has addressed the use of machine learning methods for classification at the sub-pixel level. To close this knowledge gap, in this article, six machine learning methods were compared for the specific task of sub-pixel land-cover extraction in the spatially heterogeneous region of Flanders (Belgium). In addition to the classification accuracy at the pixel and the municipality level, three evaluation criteria reflecting the methods' ease-of-use were added to the comparison: the time needed for training, the number of meta-parameters, and the minimum training set size. Robustness to changing training data was also included as the sixth evaluation criterion. Based on their scores for these six criteria, the machine learning methods were ranked according to three multi-criteria ranking scenarios. These ranking scenarios correspond to different decision-making scenarios that differ in their weighting of the criteria. In general, no overall winner could be designated: no method performs best for all evaluation scenarios. However, when both time available for preprocessing and the magnitude of the training data set are unconstrained, Support Vector Machines (SVMs) clearly outperform the other methods.

关键词： hyperspectral data   vegetation status   random forest   PROSAIL   crop  

摘要： The machine learning method, random forest (RF), is applied in order to derive biophysical and structural vegetation parameters from hyperspectral signatures. Hyperspectral data are, among other things, characterized by their high dimensionality and autocorrelation. Common multivariate regression approaches, which usually include only a limited number of spectral indices as predictors, do not make full use of the available information. In contrast, machine learning methods, such as RF, are supposed to be better suited to extract information on vegetation status. First, vegetation parameters are extracted from hyperspectral signatures simulated with the radiative transfer model, PROSAIL. Second, the transferability of these results with respect to laboratory and field measurements is investigated. In situ observations of plant physiological parameters and corresponding spectra are gathered in the laboratory for summer barley (Hordeum vulgare). Field in situ measurements focus on winter crops over several growing seasons. Chlorophyll content, Leaf Area Index and phenological growth stages are derived from simulated and measured spectra. RF performs very robustly and with a very high accuracy on PROSAIL simulated data. Furthermore, it is almost unaffected by introduced noise and bias in the data. When applied to laboratory data, the prediction accuracy is still good (C-ab: R-2 = 0.94/ LAI: R-2 = 0.80/BBCH (Growth stages of mono-and dicotyledonous plants) : R-2 = 0.91), but not as high as for simulated spectra. Transferability to field measurements is given with prediction levels as high as for laboratory data (C-ab: R-2 = 0.89/LAI: R-2 = 0.89/BBCH: R-2 = similar to 0.8). Wavelengths for deriving plant physiological status based on simulated and measured hyperspectral signatures are mostly selected from appropriate spectral regions (both field and laboratory): 700-800 nm regressing on C-ab and 800-1300 nm regressing on LAI. Results suggest that the prediction accuracy o

关键词： financial statement fraud   fraud risk assessment   fraud risk factors   machine learning   rule-based system  

摘要： This study presents a method of assessing financial statement fraud risk. The proposed approach comprises a system of financial and non-financial risk factors, and a hybrid assessment method that combines machine learning methods with a rule-based system. Experiments are performed using data from Chinese companies by four classifiers (logistic regression, back-propagation neural network, C5.0 decision tree and support vector machine) and an ensemble of those classifiers. The proposed ensemble of classifiers outperform each of the four classifiers individually in accuracy and composite error rate. The experimental results indicate that non-financial risk factors and a rule-based system help decrease the error rates. The proposed approach outperforms machine learning methods in assessing the risk of financial statement fraud. Copyright (c) 2014 John Wiley & Sons, Ltd.

关键词： Bayes Theorem   Databases, Genetic   Humans   Models, Molecular   Protein Binding   Protein Interaction Mapping/methods   Proteins/chemistry   Proteins/metabolism   Support Vector Machine   Viral Proteins/chemistry   Viral Proteins/metabolism  

摘要： Background: Viral-host protein-protein interaction plays a vital role in pathogenesis, since it defines viral infection of the host and regulation of the host proteins. Identification of key viral-host protein-protein interactions (PPIs) has great implication for therapeutics. Methods: In this study, a systematic attempt has been made to predict viral-host PPIs by integrating different features, including domain-domain association, network topology and sequence information using viral-host PPIs from VirusMINT. The three well-known supervised machine learning methods, such as SVM, Naive Bayes and Random Forest, which are commonly used in the prediction of PPIs, were employed to evaluate the performance measure based on five-fold cross validation techniques. Results: Out of 44 descriptors, best features were found to be domain-domain association and methionine, serine and valine amino acid composition of viral proteins. In this study, SVM-based method achieved better sensitivity of 67% over Naive Bayes (37.49%) and Random Forest (55.66%). However the specificity of Naive Bayes was the highest (99.52%) as compared with SVM (74%) and Random Forest (89.08%). Overall, the SVM and Random Forest achieved accuracy of 71% and 72.41%, respectively. The proposed SVM-based method was evaluated on blind dataset and attained a sensitivity of 64%, specificity of 83%, and accuracy of 74%. In addition, unknown potential targets of hepatitis B virus-human and hepatitis E virus-human PPIs have been predicted through proposed SVM model and validated by gene ontology enrichment analysis. Our proposed model shows that, hepatitis B virus "C protein'' binds to membrane docking protein, while "X protein'' and "P protein'' interacts with cell-killing and metabolic process proteins, respectively. Conclusion: The proposed method can predict large scale interspecies viral-human PPIs. The nature and function of unknown viral proteins (HBV and HEV), interacting partners of host protein were identi

关键词： artificial neural networks (ANNs)   geometric mean diameter (GMD)   parallel genetic algorithms (PGAs)   support vector machines (SVMs)  

摘要： Aggregate stability is a useful soil physical dynamic index of soil resistivity to surface wind and water erosion in all ecosystems, especially, in arid and semi-arid regions. Two machine learning techniques including support vector machines (SVMs) and artificial neural networks (ANNs) were used to develop predictive models for the estimation of geometric mean diameter (GMD) of soil aggregates. An empirical multiple linear regression (MLR) model was also constructed as the benchmark to compare their performances. Furthermore, the influence of feature space dimension reduction using parallel genetic algorithm (PGA) on the prediction accuracy of all investigated techniques was evaluated. The ANN model achieved greater accuracy in GMD prediction as compared to the MLR and SVM models. The obtained ERROR% value in GMD prediction using the ANN model was 6.9%, while it was 15.7 and 10.6% for the MLR and SVM models, respectively. Feature selection using PGA improved the prediction accuracy of all investigated techniques. The coefficient of determination (R-2) values between the measured and the predicted GMD values using PGA-based MLR, SVM, and ANN models increased by 20.0, 12.2, and 8.8% in comparison with the proposed MLR, SVM, and ANN models. In conclusion, it appears that the PGA-based ANN model could be considered as an alternative to conventional regression models for the GMD prediction.

关键词： nu-Support vector machine (nu-SVM)   Conditional value-at-risk (CVaR)   Mean-absolute semi-deviation (MASD)   Coherent measures of risk   Credit rating  

摘要： The purpose of this article is to review the similarity and difference between financial risk minimization and a class of machine learning methods known as support vector machines, which were independently developed. By recognizing their common features, we can understand them in a unified mathematical framework. On the other hand, by recognizing their difference, we can develop new methods. In particular, employing the coherent measures of risk, we develop a generalized criterion for two class classification. It includes existing criteria, such as the margin maximization and nu-SVM, as special cases. This extension can also be applied to the other type of machine learning methods such as multi-class classification, regression and outlier detection. Although the new criterion is first formulated as a nonconvex optimization, it results in a convex optimization by employing the nonnegative l(1)-regularization. Numerical examples demonstrate how the developed methods work for bond rating.

关键词： 软件安全   结合语义   机器学习   方法  

摘要： 随着计算机信息技术的不断发展,各类软件已被人们广泛应用到社会生活、生产的各个领域。但是,随着信息系统漏洞的不断增多,病毒种类的千变万化,软件安全却一直是人们感到棘手的难题。在信息技术发展的同时,近20多年新兴的机器学习也在蓬勃的发展,并且已经在很多领域得到成功的应用。本文先分析机器学习的主要策略,再分析当今软件安全的现状,最后再探索在软件安全中结合语义的机器学习方法。