目录 PrefaceChapter 1: Machine Learning ReviewMachine learning - history and definitionWhat is not machine learning?Machine learning - concepts and terminologyMachine learning - types and subtypesDatasets used imachine learningMachine learning applicationsPractical issues imachine learningMachine learning - roles and processRolesProcessMachine learning -tools and datasetsDatasetsSummaryChapter 2: Practical Approach to Real-World Supervised LearningFormal descriptioand notationData quality analysisDescriptive data analysisBasic label analysisBasic feature analysisVisualizatioanalysisUnivariate feature analysisMultivariate feature analysisData transformatioand preprocessingFeature constructionHandling missing valuesOutliersDiscretizationData samplingIs sampling needed?Undersampling and oversamplingTraining, validation, and test setFeature relevance analysis and dimensionality reductionFeature search techniquesFeature evaluatiotechniquesFilter approachWrapper approachEmbedded approachModel buildingLinear modelsLinear RegressionNaive BayesLogistic RegressionNon-linear modelsDecisioTreesK-Nearest Neiors (KNN)Support vector machines (SVM)Ensemble learning and meta learnersBootstrap aggregating or baggingBoostingModel assessment, evaluation, and comparisonsModel assessmentModel evaluatiometricsConfusiomatrix and related metricsROC and PRC curvesGaicharts and lift curvesModel comparisonsComparing two algorithmsComparing multiple algorithmsCase Study - Horse Colic ClassificationBusiness problemMachine learning mappingData analysisLabel analysisFeatures analysisSupervised learning experimentsWeka experimentsRapidMiner experimentsResults, observations, and analysisSummaryReferencesChapter 3: Unsupervised Machine Learninq Techniques……Chapter 4: Semi-Supervised and Active LearningChapter 5: Real-Time Stream Machine LearningChapter 6: Probabilistic Graph ModelingChapter 7: Deep LearningChapter 8: Text Mining and Natural Language ProcessingChapter 9: Bia Data Machine Learnina - The Final FrontierAppendix A: Linear AlgebraAppendix B: ProbabilityIndex 作者介绍 乌代?卡马特(Dr. Uday Kamath) is the chief data scientist at BAE Systems Applied Intelligence. He speizes in scalable machine learning and has spent 20 years in the domain of AML, fraud detection in finan crime, cyber security, and bioinformatics, to name a few. Dr. Kamath is responsible for key products in areas focusing on the behavioral, so networking and big data machine learning aspects of analytics at BAE AI. He received his PhD at George Mason University, under the able guidance of Dr. Kenneth De Jong, where his dissertation research focused on machine learning for big data and automated sequence mining. 序言
以下为对购买帮助不大的评价