book

Practical Machine Learning with R

by Brindha Priyadarshini Jeyaraman, Ludvig Renbo Olsen, Monicah Wambugu

August 2019

Beginner to intermediate

416 pages

7h 5m

English

Packt Publishing

Read now

Unlock full access

About the BookAbout the AuthorsDescriptionLearning ObjectivesAudienceApproachMinimum Hardware RequirementsSoftware RequirementsConventionsInstallation and SetupInstalling RInstalling R StudioInstalling LibrariesInstalling the Code BundleAdditional Resources
An Introduction to Machine LearningIntroductionThe Machine Learning ProcessRaw DataData Pre-ProcessingThe Data Splitting ProcessThe Training ProcessEvaluation ProcessDeployment ProcessProcess Flow for Making Predictions Introduction to RExercise 1: Reading from a CSV File in RStudioExercise 2: Performing Operations on a DataframeExploratory Data Analysis (EDA)View Built-in Datasets in RExercise 3: Loading Built-in DatasetsExercise 4: Viewing Summaries of DataVisualizing the DataActivity 1: Finding the Distribution of Diabetic Patients in the PimaIndiansDiabetes DatasetActivity 2: Grouping the PimaIndiansDiabetes DataActivity 3: Performing EDA on the PimaIndiansDiabetes DatasetMachine Learning ModelsTypes of PredictionSupervised LearningUnsupervised LearningApplications of Machine LearningRegressionExercise 5: Building a Linear Classifier in RActivity 4: Building Linear Models for the GermanCredit DatasetActivity 5: Using Multiple Variables for a Regression Model for the Boston Housing DatasetSummary
Data Cleaning and Pre-processingIntroductionAdvanced Operations on Data FramesExercise 6: Sorting the Data FrameJoin OperationsPre-Processing of Data FramesExercise 7: Centering VariablesExercise 8: Normalizing the VariablesExercise 9: Scaling the VariablesActivity 6: Centering and Scaling the VariablesExtracting the Principle ComponentsExercise 10: Extracting the Principle ComponentsSubsetting DataExercise 11: Subsetting a Data FrameData TransposesIdentifying the Input and Output VariablesIdentifying the Category of PredictionHandling Missing Values, Duplicates, and OutliersHandling Missing ValuesExercise 12: Identifying the Missing ValuesTechniques for Handling Missing ValuesExercise 13: Imputing Using the MICE PackageExercise 14: Performing Predictive Mean MatchingHandling DuplicatesExercise 15: Identifying DuplicatesTechniques Used to Handle Duplicate ValuesHandling OutliersExercise 16: Identifying Outlier ValuesTechniques Used to Handle OutliersExercise 17: Predicting Values to Handle OutliersHandling Missing DataExercise 18: Handling Missing ValuesActivity 7: Identifying OutliersPre-Processing Categorical DataHandling Imbalanced DatasetsUndersamplingExercise 19: Undersampling a DatasetOversamplingExercise 20: OversamplingROSEExercise 21: Oversampling using ROSESMOTEExercise 22: Implementing the SMOTE TechniqueActivity 8: Oversampling and Undersampling using SMOTEActivity 9: Sampling and Oversampling using ROSESummary
Feature EngineeringIntroductionTypes of FeaturesDatatype-Based FeaturesDate and Time FeaturesExercise 23: Creating Date FeaturesExercise 24: Creating Time FeaturesTime Series FeaturesExercise 25: BinningActivity 10: Creating Time Series Features – BinningSummary StatisticsExercise 26: Finding Description of FeaturesStandardizing and RescalingHandling Categorical VariablesSkewnessExercise 27: Computing SkewnessActivity 11: Identifying SkewnessReducing Skewness Using Log TransformExercise 28: Using Log TransformDerived Features or Domain-Specific FeaturesAdding Features to a Data FrameExercise 29: Adding a New Column to an R Data FrameHandling Redundant FeaturesExercise 30: Identifying Redundant FeaturesText Features Exercise 31: Automatically Generating Text Features Feature SelectionCorrelation AnalysisExercise 32: Plotting Correlation between Two VariablesP-ValueExercise 33: Calculating the P-ValueRecursive Feature EliminationExercise 34: Implementing Recursive Feature EliminationPCAExercise 35: Implementing PCA Activity 12: Generating PCARanking FeaturesVariable Importance Approach with Learning Vector QuantizationExercise 36: Implementing LVQVariable Importance Approach Using Random ForestsExercise 37: Finding Variable Importance in the PimaIndiansDiabetes DatasetActivity 13: Implementing the Random Forest ApproachVariable Importance Approach Using a Logistic Regression ModelExercise 38: Implementing the Logistic Regression ModelDetermining Variable Importance Using rpartExercise 39: Variable Importance Using rpart for the PimaIndiansDiabetes DataActivity 14: Selecting Features Using Variable ImportanceSummary
Introduction to neuralnet and Evaluation MethodsIntroductionClassificationBinary ClassificationExercise 40: Preparing the DatasetBalanced Partitioning Using the groupdata2 PackageExercise 41: Partitioning the DatasetExercise 42: Creating Balanced PartitionsLeakageExercise 43: Ensuring an Equal Number of Observations Per ClassStandardizingNeural Networks with neuralnetActivity 15: Training a Neural NetworkModel SelectionEvaluation MetricsAccuracyPrecisionRecallExercise 44: Creating a Confusion MatrixExercise 45: Creating Baseline EvaluationsOver and UnderfittingAdding Layers and Nodes in neuralnetCross-ValidationCreating FoldsExercise 46: Writing a Cross-Validation Training LoopActivity 16: Training and Comparing Neural Network ArchitecturesActivity 17: Training and Comparing Neural Network Architectures with Cross-ValidationMulticlass Classification OverviewSummary
Linear and Logistic Regression ModelsIntroductionRegressionLinear RegressionExercise 47: Training Linear Regression ModelsR2Exercise 48: Plotting Model PredictionsExercise 49: Incrementally Adding PredictorsComparing Linear Regression ModelsEvaluation MetricsMAERMSEDifferences between MAE and RMSEExercise 50: Comparing Models with the cvms PackageInteractionsExercise 51: Adding Interaction Terms to Our ModelShould We Standardize Predictors?Repeated Cross-ValidationExercise 52: Running Repeated Cross-ValidationExercise 53: Validating Models with validate()Activity 18: Implementing Linear RegressionLog-Transforming PredictorsExercise 54: Log-Transforming PredictorsLogistic RegressionExercise 55: Training Logistic Regression ModelsExercise 56: Creating Binomial Baseline Evaluations with cvmsExercise 57: Creating Gaussian Baseline Evaluations with cvmsRegression and Classification with Decision TreesExercise 58: Training Random Forest ModelsModel Selection by Multiple Disagreeing MetricsPareto DominanceExercise 59: Plotting the Pareto FrontActivity 19: Classifying Room TypesSummary
Unsupervised LearningIntroductionOverview of Unsupervised Learning (Clustering)Hard versus Soft ClustersFlat versus Hierarchical ClusteringMonothetic versus Polythetic ClusteringExercise 60: Monothetic and Hierarchical Clustering on a Binary DatasetDIANAExercise 61: Implement Hierarchical Clustering Using DIANAAGNESExercise 62: Agglomerative Clustering Using AGNESDistance Metrics in ClusteringExercise 63: Calculate Dissimilarity Matrices Using Euclidean and Manhattan DistanceCorrelation-Based Distance MetricsExercise 64: Apply Correlation-Based MetricsApplications of Clusteringk-means ClusteringExploratory Data Analysis Using Scatter PlotsThe Elbow MethodExercise 65: Implementation of k-means Clustering in RActivity 20: Perform DIANA, AGNES, and k-means on the Built-In Motor Car DatasetSummary
Chapter 1: An Introduction to Machine LearningActivity 1: Finding the Distribution of Diabetic Patients in the PimaIndiansDiabetes DatasetActivity 2: Grouping the PimaIndiansDiabetes DataActivity 3: Performing EDA on the PimaIndiansDiabetes DatasetActivity 4: Building Linear Models for the GermanCredit DatasetActivity 5: Using Multiple Variables for a Regression Model for the Boston Housing DatasetChapter 2: Data Cleaning and Pre-processingActivity 6: Pre-processing using Center and ScaleActivity 7: Identifying OutliersActivity 8: Oversampling and UndersamplingActivity 9: Sampling and OverSampling using ROSESolution:Chapter 3: Feature EngineeringActivity 10: Calculating Time series Feature – BinningActivity 11: Identifying SkewnessActivity 12: Generating PCAActivity 13: Implementing the Random Forest ApproachActivity 14: Selecting Features Using Variable ImportanceChapter 4: Introduction to neuralnet and Evaluation MethodsActivity 15: Training a Neural NetworkActivity 16: Training and Comparing Neural Network ArchitecturesActivity 17: Training and Comparing Neural Network Architectures with Cross-ValidationChapter 5: Linear and Logistic Regression ModelsActivity 18: Implementing Linear RegressionActivity 19: Classifying Room TypesChapter 6: Unsupervised LearningActivity 20: Perform DIANA, AGNES, and k-means on the Built-In Motor Car Dataset

Content preview from Practical Machine Learning with R

Chapter 4 Introduction to neuralnet and Evaluation Methods

Learning Objectives

By the end of this chapter, you will be able to:

Use a neural network to solve a classification problem by using the neuralnet package.
Create balanced partitions from a dataset, while decreasing leakage, with the groupdata2 package.
Evaluate and select between models using cross-validation.
Explain the concept of multiclass classification.

In this chapter, we aim to equip you with a practical understanding of neural networks.

Introduction

While neural networks have been around in some form since the mid-twentieth century, they have recently surged in popularity. Be it self-driving cars or healthcare technologies, neural networks are fundamental to some of ...