Data Science for Marketing Analytics电子书

About the Book

About the Authors

Objectives

Audience

Approach

Minimum Hardware Requirements

Software Requirements

Conventions

Installation and Setup

Installing the Code Bundle

Additional Resources

Chapter 1

Data Preparation and Cleaning

Introduction

Data Models and Structured Data

pandas

Importing and Exporting Data With pandas DataFrames

Viewing and Inspecting Data in DataFrames

Exercise 1: Importing JSON Files into pandas

Exercise 2: Identifying Semi-Structured and Unstructured Data

Structure of a pandas Series

Data Manipulation

Selecting and Filtering in pandas

Creating Test DataFrames in Python

Adding and Removing Attributes and Observations

Exercise 3: Creating and Modifying Test DataFrames

Combining Data

Handling Missing Data

Exercise 4: Combining DataFrames and Handling Missing Values

Applying Functions and Operations on DataFrames

Grouping Data

Exercise 5: Applying Data Transformations

Activity 1: Addressing Data Spilling

Summary

Chapter 2

Data Exploration and Visualization

Introduction

Identifying the Right Attributes

Exercise 6: Exploring the Attributes in Sales Data

Generating Targeted Insights

Selecting and Renaming Attributes

Transforming Values

Exercise 7: Targeting Insights for Specific Use Cases

Reshaping the Data

Exercise 8: Understanding Stacking and Unstacking

Pivot Tables

Visualizing Data

Exercise 9: Visualizing Data With pandas

Visualization through Seaborn

Visualization with Matplotlib

Activity 2: Analyzing Advertisements

Summary

Chapter 3

Unsupervised Learning: Customer Segmentation

Introduction

Customer Segmentation Methods

Traditional Segmentation Methods

Unsupervised Learning (Clustering) for Customer Segmentation

Similarity and Data Standardization

Determining Similarity

Standardizing Data

Exercise 10: Standardizing Age and Income Data of Customers

Calculating Distance

Exercise 11: Calculating Distance Between Three Customers

Activity 3: Loading, Standardizing, and Calculating Distance with a Dataset

k-means Clustering

Understanding k-means Clustering

Exercise 12: k-means Clustering on Income/Age Data

High-Dimensional Data

Exercise 13: Dealing with High-Dimensional Data

Activity 4: Using k-means Clustering on Customer Behavior Data

Summary

Chapter 4

Choosing the Best Segmentation Approach

Introduction

Choosing the Number of Clusters

Simple Visual Inspection

Exercise 14: Choosing the Number of Clusters Based on Visual Inspection

The Elbow Method with Sum of Squared Errors

Exercise 15: Determining the Number of Clusters Using the Elbow Method

Activity 5: Determining Clusters for High-End Clothing Customer Data Using the Elbow Method with the Sum of Squared Errors

Different Methods of Clustering

Mean-Shift Clustering

Exercise 16: Performing Mean-Shift Clustering to Cluster Data

k-modes and k-prototypes Clustering

Exercise 17: Clustering Data Using the k-prototypes Method

Activity 6: Using Different Clustering Techniques on Customer Behavior Data

Evaluating Clustering

Silhouette Score

Exercise 18: Calculating Silhouette Score to Pick the Best k for k-means and Comparing to the Mean-Shift Algorithm

Train and Test Split

Exercise 19: Using a Train-Test Split to Evaluate Clustering Performance

Activity 7: Evaluating Clustering on Customer Behavior Data

Summary

Chapter 5

Predicting Customer Revenue Using Linear Regression

Introduction

Understanding Regression

Feature Engineering for Regression

Feature Creation

Data Cleaning

Exercise 20: Creating Features for Transaction Data

Assessing Features Using Visualizations and Correlations

Exercise 21: Examining Relationships between Predictors and Outcome

Activity 8: Examining Relationships Between Storefront Locations and Features about Their Area

Performing and Interpreting Linear Regression

Exercise 22: Building a Linear Model Predicting Customer Spend

Activity 9: Building a Regression Model to Predict Storefront Location Revenue

Summary

Chapter 6

Other Regression Techniques and Tools for Evaluation

Introduction

Evaluating the Accuracy of a Regression Model

Residuals and Errors

Mean Absolute Error

Root Mean Squared Error

Exercise 23: Evaluating Regression Models of Location Revenue Using MAE and RMSE

Activity 10: Testing Which Variables are Important for Predicting Responses to a Marketing Offer

Using Regularization for Feature Selection

Exercise 24: Using Lasso Regression for Feature Selection

Activity 11: Using Lasso Regression to Choose Features for Predicting Customer Spend

Tree-Based Regression Models

Random Forests

Exercise 25: Using Tree-Based Regression Models to Capture Non-Linear Trends

Activity 12: Building the Best Regression Model for Customer Spend Based on Demographic Data

Summary

Chapter 7

Supervised Learning: Predicting Customer Churn

Introduction

Classification Problems

Understanding Logistic Regression

Revisiting Linear Regression

Logistic Regression

Exercise 26: Plotting the Sigmoid Function

Cost Function for Logistic Regression

Assumptions of Logistic Regression

Exercise 27: Loading, Splitting, and Applying Linear and Logistic Regression to Data

Creating a Data Science Pipeline

Obtaining the Data

Exercise 28: Obtaining the Data

Scrubbing the Data

Exercise 29: Imputing Missing Values

Exercise 30: Renaming Columns and Changing the Data Type

Exploring the Data

Statistical Overview

Correlation

Exercise 31: Obtaining the Statistical Overview and Correlation Plot

Visualizing the Data

Exercise 32: Performing Exploratory Data Analysis (EDA)

Activity 13: Performing OSE of OSEMN

Modeling the Data

Feature Selection

Exercise 33: Performing Feature Selection

Model Building

Exercise 34: Building a Logistic Regression Model

Interpreting the Data

Activity 14: Performing MN of OSEMN

Summary

Chapter 8

Fine-Tuning Classification Algorithms

Introduction

Support Vector Machines

Intuition Behind Maximum Margin

Linearly Inseparable Cases

Linearly Inseparable Cases Using Kernel

Exercise 35: Training an SVM Algorithm Over a Dataset

Decision Trees

Exercise 36: Implementing a Decision Tree Algorithm Over a Dataset

Important Terminology of Decision Trees

Decision Tree Algorithm Formulation

Random Forest

Exercise 37: Implementing a Random Forest Model Over a Dataset

Activity 15: Implementing Different Classification Algorithms

Preprocessing Data for Machine Learning Models

Standardization

Exercise 38: Standardizing Data

Scaling

Exercise 39: Scaling Data After Feature Selection

Normalization

Exercise 40: Performing Normalization on Data

Model Evaluation

Exercise 41: Implementing Stratified k-fold

Fine-Tuning of the Model

Exercise 42: Fine-Tuning a Model

Activity 16: Tuning and Optimizing the Model

Performance Metrics

Precision

Recall

F1 Score

Exercise 43: Evaluating the Performance Metrics for a Model

ROC Curve

Exercise 44: Plotting the ROC Curve

Activity 17: Comparison of the Models

Summary

Chapter 9

Modeling Customer Choice

Introduction

Understanding Multiclass Classification

Classifiers in Multiclass Classification

Exercise 45: Implementing a Multiclass Classification Algorithm on a Dataset

Performance Metrics

Exercise 46: Evaluating Performance Using Multiclass Performance Metrics

Activity 18: Performing Multiclass Classification and Evaluating Performance

Class Imbalanced Data

Exercise 47: Performing Classification on Imbalanced Data

Dealing with Class-Imbalanced Data

Exercise 48: Visualizing Sampling Techniques

Exercise 49: Fitting a Random Forest Classifier Using SMOTE and Building the Confusion Matrix

Activity 19: Dealing with Imbalanced Data

Summary

Appendix

Chapter 1: Data Preparation and Cleaning

Activity 1: Addressing Data Spilling

Chapter 2: Data Exploration and Visualization

Activity 2: Analyzing Advertisements

Chapter 3: Unsupervised Learning: Customer Segmentation

Activity 3: Loading, Standardizing, and Calculating Distance with a Dataset

Activity 4: Using k-means Clustering on Customer Behavior Data

Chapter 4: Choosing the Best Segmentation Approach

Activity 5: Determining Clusters for High-End Clothing Customer Data Using the Elbow Method with the Sum of Squared Errors

Activity 6: Using Different Clustering Techniques on Customer Behavior Data

Activity 7: Evaluating Clustering on Customer Behavior Data

Chapter 5: Predicting Customer Revenue Using Linear Regression

Activity 8: Examining Relationships between Storefront Locations and Features about their Area

Activity 9: Building a Regression Model to Predict Storefront Location Revenue

Chapter 6: Other Regression Techniques and Tools for Evaluation

Activity 10: Testing Which Variables are Important for Predicting Responses to a Marketing Offer

Activity 11: Using Lasso Regression to Choose Features for Predicting Customer Spend

Activity 12: Building the Best Regression Model for Customer Spend Based on Demographic Data

Chapter 7: Supervised Learning: Predicting Customer Churn

Activity 13: Performing OSE from OSEMN

Activity 14: Performing MN of OSEMN

Chapter 8: Fine-Tuning Classification Algorithms

Activity 15: Implementing Different Classification Algorithms

Activity 16: Tuning and Optimizing the Model

Activity 17: Comparison of the Models

Chapter 9: Modeling Customer Choice

Activity 18: Performing Multiclass Classification and Evaluating Performance

Activity 19: Dealing with Imbalanced Data