Data Science Using Python and R

Learn data science by doing data science! 

Data Science Using Python and R will get you plugged into the world’s two most widespread open-source platforms for data science: Python and R.

Data science is hot. Bloomberg called data scientist “the hottest job in America.” Python and R are the top two open-source data science tools in the world. In Data Science Using Python and R, you will learn step-by-step how to produce hands-on solutions to real-world business problems, using state-of-the-art techniques. 

Data Science Using Python and R is written for the general reader with no previous analytics or programming experience. An entire chapter is dedicated to learning the basics of Python and R. Then, each chapter presents step-by-step instructions and walkthroughs for solving data science problems using Python and R.

Those with analytics experience will appreciate having a one-stop shop for learning how to do data science using Python and R. Topics covered include data preparation, exploratory data analysis, preparing to model the data, decision trees, model evaluation, misclassification costs, naïve Bayes classification, neural networks, clustering, regression modeling, dimension reduction, and association rules mining.

Further, exciting new topics such as random forests and general linear models are also included. The book emphasizes data-driven error costs to enhance profitability, which avoids the common pitfalls that may cost a company millions of dollars.

Data Science Using Python and R provides exercises at the end of every chapter, totaling over 500 exercises in the book. Readers will therefore have plenty of opportunity to test their newfound data science skills and expertise. In the Hands-on Analysis exercises, readers are challenged to solve interesting business problems using real-world data sets.

CHANTAL D. LAROSE, PHD, is an Assistant Professor of Statistics & Data Science at Eastern Connecticut State University (ECSU). She has co-authored three books on data science and predictive analytics and helped develop data science programs at ECSU and SUNY New Paltz. Her PhD dissertation, Model-Based Clustering of Incomplete Data, tackles the persistent problem of trying to do data science with incomplete data.

DANIEL T. LAROSE, PHD, is a Professor of Data Science and Statistics and Director of the Data Science programs at Central Connecticut State University. He has published many books on data science, data mining, predictive analytics, and statistics. His consulting clients include The Economist magazine, Forbes Magazine, the CIT Group, and Microsoft.

Preface xi

About the Authors xv

Acknowledgements xvii

Chapter 1 Introduction to Data Science 1

1.1 Why Data Science? 1

1.2 What is Data Science? 1

1.3 The Data Science Methodology 2

1.4 Data Science Tasks 5

1.4.1 Description 6

1.4.2 Estimation 6

1.4.3 Classification 6

1.4.4 Clustering 7

1.4.5 Prediction 7

1.4.6 Association 7

Exercises 8

Chapter 2 The Basics of Python and R 9

2.1 Downloading Python 9

2.2 Basics of Coding in Python 9

2.2.1 Using Comments in Python 9

2.2.2 Executing Commands in Python 10

2.2.3 Importing Packages in Python 11

2.2.4 Getting Data into Python 12

2.2.5 Saving Output in Python 13

2.2.6 Accessing Records and Variables in Python 14

2.2.7 Setting Up Graphics in Python 15

2.3 Downloading R and RStudio 17

2.4 Basics of Coding in R 19

2.4.1 Using Comments in R 19

2.4.2 Executing Commands in R 20

2.4.3 Importing Packages in R 20

2.4.4 Getting Data into R 21

2.4.5 Saving Output in R 23

2.4.6 Accessing Records and Variables in R 24

References 26

Exercises 26

Chapter 3 Data Preparation 29

3.1 The Bank Marketing Data Set 29

3.2 The Problem Understanding Phase 29

3.2.1 Clearly Enunciate the Project Objectives 29

3.2.2 Translate These Objectives into a Data Science Problem 30

3.3 Data Preparation Phase 31

3.4 Adding an Index Field 31

3.4.1 How to Add an Index Field Using Python 31

3.4.2 How to Add an Index Field Using R 32

3.5 Changing Misleading Field Values 33

3.5.1 How to Change Misleading Field Values Using Python 34

3.5.2 How to Change Misleading Field Values Using R 34

3.6 Reexpression of Categorical Data as Numeric 36

3.6.1 How to Reexpress Categorical Field Values Using Python 36

3.6.2 How to Reexpress Categorical Field Values Using R 38

3.7 Standardizing the Numeric Fields 39

3.7.1 How to Standardize Numeric Fields Using Python 40

3.7.2 How to Standardize Numeric Fields Using R 40

3.8 Identifying Outliers 40

3.8.1 How to Identify Outliers Using Python 41

3.8.2 How to Identify Outliers Using R 42

References 43

Exercises 44

Chapter 4 Exploratory Data Analysis 47

4.1 EDA Versus HT 47

4.2 Bar Graphs with Response Overlay 47

4.2.1 How to Construct a Bar Graph with Overlay Using Python 49

4.2.2 How to Construct a Bar Graph with Overlay Using R 50

4.3 Contingency Tables 51

4.3.1 How to Construct Contingency Tables Using Python 52

4.3.2 How to Construct Contingency Tables Using R 53

4.4 Histograms with Response Overlay 53

4.4.1 How to Construct Histograms with Overlay Using Python 55

4.4.2 How to Construct Histograms with Overlay Using R 58

4.5 Binning Based on Predictive Value 58

4.5.1 How to Perform Binning Based on Predictive Value Using Python 59

4.5.2 How to Perform Binning Based on Predictive Value Using R 62

References 63

Exercises 63

Chapter 5 Preparing to Model the Data 69

5.1 The Story So Far 69

5.2 Partitioning the Data 69

5.2.1 How to Partition the Data in Python 70

5.2.2 How to Partition the Data in R 71

5.3 Validating your Partition 72

5.4 Balancing the Training Data Set 73

5.4.1 How to Balance the Training Data Set in Python 74

5.4.2 How to Balance the Training Data Set in R 75

5.5 Establishing Baseline Model Performance 77

References 78

Exercises 78

Chapter 6 Decision Trees 81

6.1 Introduction to Decision Trees 81

6.2 Classification and Regression Trees 83

6.2.1 How to Build CART Decision Trees Using Python 84

6.2.2 How to Build CART Decision Trees Using R 86

6.3 The C5.0 Algorithm for Building Decision Trees 88

6.3.1 How to Build C5.0 Decision Trees Using Python 89

6.3.2 How to Build C5.0 Decision Trees Using R 90

6.4 Random Forests 91

6.4.1 How to Build Random Forests in Python 92

6.4.2 How to Build Random Forests in R 92

References 93

Exercises 93

Chapter 7 Model Evaluation 97

7.1 Introduction to Model Evaluation 97

7.2 Classification Evaluation Measures 97

7.3 Sensitivity and Specificity 99

7.4 Precision, Recall, and F_β Scores 99

7.5 Method for Model Evaluation 100

7.6 An Application of Model Evaluation 100

7.6.1 How to Perform Model Evaluation Using R 103

7.7 Accounting for Unequal Error Costs 104

7.7.1 Accounting for Unequal Error Costs Using R 105

7.8 Comparing Models with and without Unequal Error Costs 106

7.9 Data‐Driven Error Costs 107

Exercises 109

Chapter 8 Naïve Bayes Classification 113

8.1 Introduction to Naive Bayes 113

8.2 Bayes Theorem 113

8.3 Maximum a Posteriori Hypothesis 114

8.4 Class Conditional Independence 114

8.5 Application of Naive Bayes Classification 115

8.5.1 Naive Bayes in Python 121

8.5.2 Naive Bayes in R 123

References 125

Exercises 126

Chapter 9 Neural Networks 129

9.1 Introduction to Neural Networks 129

9.2 The Neural Network Structure 129

9.3 Connection Weights and the Combination Function 131

9.4 The Sigmoid Activation Function 133

9.5 Backpropagation 134

9.6 An Application of a Neural Network Model 134

9.7 Interpreting the Weights in a Neural Network Model 136

9.8 How to Use Neural Networks in R 137

References 138

Exercises 138

Chapter 10 Clustering 141

10.1 What is Clustering? 141

10.2 Introduction to the K‐Means Clustering Algorithm 142

10.3 An Application of K‐Means Clustering 143

10.4 Cluster Validation 144

10.5 How to Perform K‐Means Clustering Using Python 145

10.6 How to Perform K‐Means Clustering Using R 147

Exercises 149

Chapter 11 Regression Modeling 151

11.1 The Estimation Task 151

11.2 Descriptive Regression Modeling 151

11.3 An Application of Multiple Regression Modeling 152

11.4 How to Perform Multiple Regression Modeling Using Python 154

11.5 How to Perform Multiple Regression Modeling Using R 156

11.6 Model Evaluation for Estimation 157

11.6.1 How to Perform Estimation Model Evaluation Using Python 159

11.6.2 How to Perform Estimation Model Evaluation Using R 160

11.7 Stepwise Regression 161

11.7.1 How to Perform Stepwise Regression Using R 162

11.8 Baseline Models for Regression 162

References 163

Exercises 164

Chapter 12 Dimension Reduction 167

12.1 The Need for Dimension Reduction 167

12.2 Multicollinearity 168

12.3 Identifying Multicollinearity Using Variance Inflation Factors 171

12.3.1 How to Identify Multicollinearity Using Python 172

12.3.2 How to Identify Multicollinearity in R 173

12.4 Principal Components Analysis 175

12.5 An Application of Principal Components Analysis 175

12.6 How Many Components Should We Extract? 176

12.6.1 The Eigenvalue Criterion 176

12.6.2 The Proportion of Variance Explained Criterion 177

12.7 Performing Pca with K = 4 178

12.8 Validation of the Principal Components 178

12.9 How to Perform Principal Components Analysis Using Python 179

12.10 How to Perform Principal Components Analysis Using R 181

12.11 When is Multicollinearity Not a Problem? 183

References 184

Exercises 184

Chapter 13 Generalized Linear Models 187

13.1 An Overview of General Linear Models 187

13.2 Linear Regression as a General Linear Model 188

13.3 Logistic Regression as a General Linear Model 188

13.4 An Application of Logistic Regression Modeling 189

13.4.1 How to Perform Logistic Regression Using Python 190

13.4.2 How to Perform Logistic Regression Using R 191

13.5 Poisson Regression 192

13.6 An Application of Poisson Regression Modeling 192

13.6.1 How to Perform Poisson Regression Using Python 193

13.6.2 How to Perform Poisson Regression Using R 194

Reference 195

Exercises 195

Chapter 14 Association Rules 199

14.1 Introduction to Association Rules 199

14.2 A Simple Example of Association Rule Mining 200

14.3 Support, Confidence, and Lift 200

14.4 Mining Association Rules 202

14.4.1 How to Mine Association Rules Using R 203

14.5 Confirming Our Metrics 207

14.6 The Confidence Difference Criterion 208

14.6.1 How to Apply the Confidence Difference Criterion Using R 208

14.7 The Confidence Quotient Criterion 209

14.7.1 How to Apply the Confidence Quotient Criterion Using R 210

References 211

Exercises 211

Appendix Data Summarization and Visualization 215

Part 1: Summarization 1: Building Blocks of Data Analysis 215

Part 2: Visualization: Graphs and Tables for Summarizing and Organizing Data 217

Part 3: Summarization 2: Measures of Center, Variability, and Position 222

Part 4: Summarization and Visualization of Bivariate Elationships 225

Index 231 

The New copy of this book will include any supplemental materials advertised. Please check the title of the book to determine if it should include any access cards, study guides, lab manuals, CDs, etc.

The Used, Rental and eBook copies of this book are not guaranteed to include any supplemental materials. Typically, only the book itself is included. This is true even if the title states it includes any access cards, study guides, lab manuals, CDs, etc.