book

Applied Supervised Learning with R

Name: Applied Supervised Learning with R
ISBN: 9781838556334

by Karthik Ramasubramanian, Jojo Moolayil

May 2019

Intermediate to advanced

502 pages

9h 44m

English

Packt Publishing

Read now

Unlock full access

Preface
About the BookAbout the AuthorsLearning ObjectivesAudienceApproachMinimum Hardware RequirementsSoftware RequirementsConventionsInstallation and SetupInstalling the Code BundleAdditional Resources
Chapter 1:
R for Advanced AnalyticsIntroductionWorking with Real-World DatasetsExercise 1: Using the unzip Method for Unzipping a Downloaded FileReading Data from Various Data FormatsCSV FilesExercise 2: Reading a CSV File and Summarizing its ColumnJSONExercise 3: Reading a JSON file and Storing the Data in DataFrameTextExercise 4: Reading a CSV File with Text Column and Storing the Data in VCorpusWrite R Markdown Files for Code ReproducibilityActivity 1: Create an R Markdown File to Read a CSV File and Write a Summary of DataData Structures in RVectorMatrixExercise 5: Performing Transformation on the Data to Make it Available for the AnalysisListExercise 6: Using the List Method for Storing Integers and Characters TogetherActivity 2: Create a List of Two Matrices and Access the ValuesDataFrameExercise 7: Performing Integrity Checks Using DataFrameData TableExercise 8: Exploring the File Read OperationData Processing and TransformationcbindExercise 9: Exploring the cbind FunctionrbindExercise 10: Exploring the rbind FunctionThe merge FunctionExercise 11: Exploring the merge FunctionInner JoinLeft JoinRight JoinFull JoinThe reshape FunctionExercise 12: Exploring the reshape FunctionThe aggregate FunctionThe Apply Family of FunctionsThe apply FunctionExercise 13: Implementing the apply FunctionThe lapply FunctionExercise 14: Implementing the lapply FunctionThe sapply FunctionThe tapply FunctionUseful PackagesThe dplyr PackageExercise 15: Implementing the dplyr PackageThe tidyr PackageExercise 16: Implementing the tidyr PackageActivity 3: Create a DataFrame with Five Summary Statistics for All Numeric Variables from Bank Data Using dplyr and tidyrThe plyr PackageExercise 17: Exploring the plyr PackageThe caret PackageData VisualizationScatterplotScatter Plot between Age and Balance split by Marital StatusLine ChartsHistogramBoxplotSummary
Chapter 2:
Exploratory Analysis of DataIntroductionDefining the Problem StatementProblem-Designing ArtifactsUnderstanding the Science Behind EDAExploratory Data AnalysisExercise 18: Studying the Data DimensionsUnivariate AnalysisExploring Numeric/Continuous FeaturesExercise 19: Visualizing Data Using a Box PlotExercise 20: Visualizing Data Using a HistogramExercise 21: Visualizing Data Using a Density PlotExercise 22: Visualizing Multiple Variables Using a HistogramActivity 4: Plotting Multiple Density Plots and BoxplotsExercise 23: Plotting a Histogram for the nr.employed, euribor3m, cons.conf.idx, and duration VariablesExploring Categorical FeaturesExercise 24: Exploring Categorical FeaturesExercise 25: Exploring Categorical Features Using a Bar ChartExercise 26: Exploring Categorical Features using Pie ChartExercise 27: Automate Plotting Categorical VariablesExercise 28: Automate Plotting for the Remaining Categorical VariablesExercise 29: Exploring the Last Remaining Categorical Variable and the Target VariableBivariate AnalysisStudying the Relationship between Two Numeric VariablesExercise 30: Studying the Relationship between Employee Variance Rate and Number of EmployeesStudying the Relationship between a Categorical and a Numeric VariableExercise 31: Studying the Relationship between the y and age VariablesExercise 32: Studying the Relationship between the Average Value and the y VariableExercise 33: Studying the Relationship between the cons.price.idx, cons.conf.idx, curibor3m, and nr.employed VariablesStudying the Relationship Between Two Categorical VariablesExercise 34: Studying the Relationship Between the Target y and marital status VariablesExercise 35: Studying the Relationship between the job and education VariablesMultivariate AnalysisValidating Insights Using Statistical TestsCategorical Dependent and Numeric/Continuous Independent VariablesExercise 36: Hypothesis 1 Testing for Categorical Dependent Variables and Continuous Independent VariablesExercise 37: Hypothesis 2 Testing for Categorical Dependent Variables and Continuous Independent VariablesCategorical Dependent and Categorical Independent VariablesExercise 38: Hypothesis 3 Testing for Categorical Dependent Variables and Categorical Independent VariablesExercise 39: Hypothesis 4 and 5 Testing for a Categorical Dependent Variable and a Categorical Independent VariableCollating Insights – Refine the Solution to the ProblemSummary
Chapter 3:
Introduction to Supervised LearningIntroductionSummary of the Beijing PM2.5 DatasetExercise 40: Exploring the DataRegression and Classification ProblemsMachine Learning WorkflowDesign the ProblemSource and Prepare DataCode the ModelTrain and EvaluateExercise 41: Creating a Train-and-Test Dataset Randomly Generated by the Beijing PM2.5 DatasetDeploy the ModelRegressionSimple and Multiple Linear RegressionAssumptions in Linear Regression ModelsExploratory Data Analysis (EDA)Exercise 42: Exploring the Time Series Views of PM2.5, DEWP, TEMP, and PRES variables of the Beijing PM2.5 DatasetExercise 43: Undertaking Correlation AnalysisExercise 44: Drawing a Scatterplot to Explore the Relationship between PM2.5 Levels and Other FactorsActivity 5: Draw a Scatterplot between PRES and PM2.5 Split by MonthsModel BuildingExercise 45: Exploring Simple and Multiple Regression ModelsModel InterpretationClassificationLogistic RegressionA Brief IntroductionMechanics of Logistic RegressionModel BuildingExercise 46: Storing the Rolling 3-Hour Average in the Beijing PM2.5 DatasetActivity 6: Transforming Variables and Deriving New Variables to Build a ModelInterpreting a ModelEvaluation MetricsMean Absolute Error (MAE)Root Mean Squared Error (RMSE)R-squaredAdjusted R-squareMean Reciprocal Rank (MRR)Exercise 47: Finding Evaluation MetricsConfusion Matrix-Based MetricsAccuracySensitivitySpecificityF1 ScoreExercise 48: Working with Model Evaluation on Training DataReceiver Operating Characteristic (ROC) CurveExercise 49: Creating an ROC CurveSummary
Chapter 4:
RegressionIntroductionLinear RegressionExercise 50: Print the Coefficient and Residual Values Using the multiple_PM_25_linear_model ObjectActivity 7: Printing Various Attributes Using Model Object Without Using the Summary FunctionExercise 51: Add the Interaction Term DEWP:TEMP:month in the lm() FunctionModel DiagnosticsExercise 52: Generating and Fitting Models Using the Linear and Quadratic EquationsResidual versus Fitted PlotNormal Q-Q PlotScale-Location PlotResidual versus LeverageImproving the ModelTransform the Predictor or Target VariableChoose a Non-Linear ModelRemove an Outlier or Influential PointAdding the Interaction EffectQuantile RegressionExercise 53: Fit a Quantile Regression on the Beijing PM2.5 DatasetExercise 54: Plotting Various Quantiles with More GranularityPolynomial RegressionExercise 55: Performing Uniform Distribution Using the runif() FunctionRidge RegressionRegularization Term – L2 NormExercise 56: Ridge Regression on the Beijing PM2.5 datasetLASSO RegressionExercise 57: LASSO RegressionElastic Net RegressionExercise 58: Elastic Net RegressionComparison between Coefficients and Residual Standard ErrorExercise 59: Computing the RSE of Linear, Ridge, LASSO, and Elastic Net RegressionsPoisson RegressionExercise 60: Performing Poisson RegressionExercise 61: Computing OverdispersionCox Proportional-Hazards Regression ModelNCCTG Lung Cancer DataExercise 62: Exploring the NCCTG Lung Cancer Data Using Cox-RegressionSummary
Chapter 5:
ClassificationIntroductionGetting Started with the Use CaseSome Background on the Use CaseDefining the Problem StatementData GatheringExercise 63: Exploring Data for the Use CaseExercise 64: Calculating the Null Value Percentage in All ColumnsExercise 65: Removing Null Values from the DatasetExercise 66: Engineer Time-Based Features from the Date VariableExercise 67: Exploring the Location FrequencyExercise 68: Engineering the New Location with Reduced LevelsClassification Techniques for Supervised LearningLogistic RegressionHow Does Logistic Regression Work?Exercise 69: Build a Logistic Regression ModelInterpreting the Results of Logistic RegressionEvaluating Classification ModelsConfusion Matrix and Its Derived MetricsWhat Metric Should You Choose?Evaluating Logistic RegressionExercise 70: Evaluate a Logistic Regression ModelExercise 71: Develop a Logistic Regression Model with All of the Independent Variables Available in Our Use CaseActivity 8: Building a Logistic Regression Model with Additional FeaturesDecision TreesHow Do Decision Trees Work?Exercise 72: Create a Decision Tree Model in RActivity 9: Create a Decision Tree Model with Additional Control ParametersEnsemble ModellingRandom ForestWhy Are Ensemble Models Used?Bagging – Predecessor to Random ForestHow Does Random Forest Work?Exercise 73: Building a Random Forest Model in RActivity 10: Build a Random Forest Model with a Greater Number of TreesXGBoostHow Does the Boosting Process Work?What Are Some Popular Boosting Techniques?How Does XGBoost Work?Implementing XGBoost in RExercise 74: Building an XGBoost Model in RExercise 75: Improving the XGBoost Model's PerformanceDeep Neural NetworksA Deeper Look into Deep Neural NetworksHow Does the Deep Learning Model Work?What Framework Do We Use for Deep Learning Models?Building a Deep Neural Network in KerasExercise 76: Build a Deep Neural Network in R using R KerasChoosing the Right Model for Your Use CaseSummary
Chapter 6:
Feature Selection and Dimensionality ReductionIntroductionFeature EngineeringDiscretizationExercise 77: Performing Binary DiscretizationMulti-Category DiscretizationExercise 78: Demonstrating the Use of Quantile FunctionOne-Hot EncodingExercise 79: Using One-Hot EncodingActivity 11: Converting the CBWD Feature of the Beijing PM2.5 Dataset into One-Hot Encoded ColumnsLog TransformationExercise 80: Performing Log TransformationFeature SelectionUnivariate Feature SelectionExercise 81: Exploring Chi-SquaredHighly Correlated VariablesExercise 82: Plotting a Correlated MatrixModel-Based Feature Importance RankingExercise 83: Exploring RFE Using RFExercise 84: Exploring the Variable Importance using the Random Forest ModelFeature ReductionPrincipal Component Analysis (PCA)Exercise 85: Performing PCAVariable ClusteringExercise 86: Using Variable ClusteringLinear Discriminant Analysis for Feature ReductionExercise 87: Exploring LDASummary
Chapter 7:
Model ImprovementsIntroductionBias-Variance Trade-offWhat is Bias and Variance in Machine Learning Models?Underfitting and OverfittingDefining a Sample Use CaseExercise 88: Loading and Exploring DataCross-ValidationHoldout Approach/ValidationExercise 89: Performing Model Assessment Using Holdout ValidationK-Fold Cross-ValidationExercise 90: Performing Model Assessment Using K-Fold Cross-ValidationHold-One-Out ValidationExercise 91: Performing Model Assessment Using Hold-One-Out ValidationHyperparameter OptimizationGrid Search OptimizationExercise 92: Performing Grid Search Optimization – Random ForestExercise 93: Grid Search Optimization – XGBoostRandom Search OptimizationExercise 94: Using Random Search Optimization on a Random Forest ModelExercise 95: Random Search Optimization – XGBoostBayesian OptimizationExercise 96: Performing Bayesian Optimization on the Random Forest ModelExercise 97: Performing Bayesian Optimization using XGBoostActivity 12: Performing Repeated K-Fold Cross Validation and Grid Search OptimizationSummary
Chapter 8:
Model DeploymentIntroductionWhat is an API?Introduction to plumberExercise 98: Developing an ML Model and Deploying It as a Web Service Using PlumberChallenges in Deploying Models with plumberA Brief History of the Pre-Docker EraDockerDeploying the ML Model Using Docker and plumberExercise 99: Create a Docker Container for the R plumber ApplicationDisadvantages of Using plumber to Deploy R ModelsAmazon Web ServicesIntroducing AWS SageMakerDeploying an ML Model Endpoint Using SageMakerExercise 100: Deploy the ML Model as a SageMaker EndpointWhat is Amazon Lambda?What is Amazon API Gateway?Building Serverless ML ApplicationsExercise 101: Building a Serverless Application Using API Gateway, AWS Lambda, and SageMakerDeleting All Cloud Resources to Stop BillingActivity 13: Deploy an R Model Using plumberSummary
Chapter 9:
Capstone Project - Based on Research PapersIntroductionExploring Research WorkThe mlr PackageOpenML PackageProblem Design from the Research PaperFeatures in Scene DatasetImplementing Multilabel Classifier Using the mlr and OpenML PackagesExercise 102: Downloading the Scene Dataset from OpenMLConstructing a LearnerAdaptation MethodsTransformation MethodsBinary Relevance MethodClassifier Chains MethodNested StackingDependent Binary Relevance MethodStackingExercise 103: Generating Decision Tree Model Using the classif.rpart MethodTrain the ModelExercise 104: Train the ModelPredicting the OutputPerformance of the ModelResampling the DataBinary Performance for Each LabelBenchmarking ModelConducting Benchmark ExperimentsExercise 105: Exploring How to Conduct a Benchmarking on Various LearnersAccessing Benchmark ResultsLearner PerformancesPredictionsLearners and measuresActivity 14: Getting the Binary Performance Step with classif.C50 Learner Instead of classif.rpartWorking with OpenML Upload FunctionsSummary

Appendix
Chapter 1: R for Advanced AnalyticsActivity 1: Create an R Markdown File to Read a CSV File and Write a Summary of DataActivity 2: Create a List of Two Matrices and Access the ValuesActivity 3: Create a DataFrame with Five Summary Statistics for All Numeric Variables from Bank Data Using dplyr and tidyrChapter 2: Exploratory Analysis of DataActivity 4: Plotting Multiple Density Plots and BoxplotsChapter 3: Introduction to Supervised LearningActivity 5: Draw a Scatterplot between PRES and PM2.5 Split by MonthsActivity 6: Transforming Variables and Deriving New Variables to Build a ModelChapter 4: RegressionActivity 7: Printing Various Attributes Using Model Object Without Using the summary FunctionChapter 5: ClassificationActivity 8: Building a Logistic Regression Model with Additional FeaturesActivity 9: Create a Decision Tree Model with Additional Control ParametersActivity 10: Build a Random Forest Model with a Greater Number of TreesChapter 6: Feature Selection and Dimensionality ReductionActivity 11: Converting the CBWD Feature of the Beijing PM2.5 Dataset into One-Hot Encoded ColumnsChapter 7: Model ImprovementsActivity 12: Perform Repeated K-Fold Cross Validation and Grid Search OptimizationChapter 8: Model DeploymentActivity 13: Deploy an R Model Using PlumberChapter 9: Capstone Project - Based on Research PapersActivity 14: Getting the Binary Performance Step with classif.C50 Learner Instead of classif.rpart

Content preview from Applied Supervised Learning with R

Chapter 3:

Introduction to Supervised Learning

Learning Objectives

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

Explain supervised learning and machine learning workflow
Use and explore the Beijing PM2.5 dataset
Explain the difference between continuous and categorical dependent variables
Implement the basic regression and classification algorithms in R
Identify the key differences between supervised learning and other types of machine learning
Work with the evaluation metrics of supervised learning algorithms
Perform model diagnostics for avoiding biased coefficient estimates and large standard errors

In this chapter, we will introduce supervised learning and demonstrate the workflow of building machine learning models with real-world ...

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

Read now

Unlock full access

More than 5,000 organizations count on O’Reilly

O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.

Julian F.

Head of Cybersecurity

I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.

Addison B.

Field Engineer

I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.

Amir M.

Data Platform Tech Lead

I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.

Mark W.

Embedded Software Engineer

Publisher Resources

ISBN: 9781838556334

Cloud Computing

Data Engineering

Data Science

AI & ML

Programming Languages

Software Architecture

IT/Ops

Security

Design