Machine Learning for Marketing

by Hiroshi Mamitsuka

Overview

Machine learning, now a central part of artificial intelligence, would be a driving force to change the current world to a more autonomous society. This impact of machine learning appears in many fields, for example, science, engineering, finance, agriculture, to name a few. Marketing is rather behind this trend, while marketing has a lot of potential applications for machine learning. In other words, marketing may change into more autonomous scientific work by using data and also proper formulation of each application into a machine learning problem. This book focuses on two major, traditional paradigms of marketing: target marketing and relationship marketing. Then it is revealed that each of numerous aspects of the two marketing paradigms can be formulated into a machine learning problem. That is, for each problem, a machine learning model can be built and parameters of the model can be estimated/optimized from given data. For example, an important objective of target marketing can be interpreted as a problem of finding a customer segment, which has a plenty of customers but no competitors. This problem can be formulated into a machine learning problem for which a model is built and model parameters can be estimated from given data. This book, for each machine learning problem setting, always builds a simpler (probably simplest) model, so that readers can understand the idea and assumption of the model easily. This book would be useful for both sides of marketing and machine learning. That is, marketers would be able to study the way of formulating a problem of marketing into a machine learning problem/function in which parameters are estimated from given data. On the other hand, machine learners would be able to study applications of marketing and also essential and intuitive ideas behind marketing through numerous applications in this book.

Product Details

ISBN-13:	9784991044526
Publisher:	Global Data Science Publishing
Publication date:	06/14/2019
Pages:	238
Product dimensions:	6.14(w) x 9.21(h) x 0.50(d)

Table of Contents

1 Introduction

2 Concepts and Terminology

2.1 Data Types

2.2 Machine Learning (ML)

2.3 Marketing

3 Target Marketing

3.1 Market Segmentation

3.2 Market Targeting

3.2.1 Segment Evaluation

3.2.2 Segment Selection

3.3 Market Positioning

3.3.1 Identifying Competitors: SWOT Analysis

3.3.2 Understanding Competitors: Perceptual Map

4 Relationship Marketing

4.1 Customer Relationship Management: Idea and Objective

4.1.1 Protable Customer

4.1.2 Extracting Protable Customers - RFM Analysis

4.1.3 Customer Loyalty Satisfaction

4.1.4 Customer Lifetime Value (CLV)

4.2 Retention Marketing

4.2.1 Understanding Customers: Marketing Funnel

4.2.2 Boosting Customer Loyalty

4.3 Marketing Communications

4.3.1 Types of Communication Channels

4.3.2 Capturing Data through Communication Channels

5 ML for Marketing

5.1 Organization of this Chapter

5.1.1 Regular Machine Learning

5.1.2 Feature Learning

5.1.3 Kernel Learning

5.2 Regular Machine Learning

5.2.1 Learning Vectors

5.2.2 Learning Nodes in a Graph

5.2.3 Learning Multiple Data Types: Data Integrative Learning

5.3 Feature Learning

5.3.1 Feature Selection

5.3.2 Feature Generation

5.4 Kernel Learning/ Kernel Method

5.4.1 Kernel Function

5.4.2 Kernel Learning/ Kernel Method

5.4.3 Kernel Method Example: Support Vector Machine

5.4.4 Pros and Cons of Kernel Learning

5.5 Key Points

5.5.1 Squared Distance and L2 Norm

5.5.2 Rayleigh Quotient

6 ML for Target Marketing

6.1 Data

6.1.1 Customers

6.1.2 Products

6.2 Market Segmentation

6.2.1 Clustering (Unsupervised Setting)

6.2.2 Classication (Supervised Setting)

6.3 Market Targeting

6.3.1 Evaluating Segments

6.3.2 Selecting Segments Using Customer Features

6.4 Market Positioning

6.4.1 SWOT Analysis

6.4.2 Generating Perceptual Map

6.4.3 Finding the Suitable Place for a New Product against Competitors in Market

6.5 Summary of Machine Learning for Target Marketing

6.5.1 Data Space on Customers and Products

7 ML for Relationship Marketing

7.1 Customer Relationship Management

7.1.1 Learning (Customers Products)-Matrix

7.1.2 Detecting Protable Customers

7.1.3 Customer Loyalty Satisfaction and Customer Lifetime Value

7.2 Retention Marketing: Learning Marketing Funnel

7.2.1 Data Assumption

7.2.2 Data Acquisition

7.2.3 Problem Setting

7.2.4 Discrete Approach

7.2.5 Probabilistic Approach

7.3 Market Communications

7.3.1 Optimizing Market Communication Channels

8 ML for Database Marketing

8.1 Matrices (without Similarity Matrices) Sharing Instances

8.1.1 Supervised Learning

8.1.2 Unsupervised Learning: Clustering

8.1.3 Unsupervised Learning: Matrix Factorization (Collaborative Matrix Factorization)

8.1.4 Examining Interactions between Features

8.1.5 Notes

8.2 Matrices (with Similarity Matrices) Sharing Instances

8.2.1 Supervised Learning

8.2.2 Unsupervised Learning: Clustering

8.2.3 Unsupervised Learning: Collaborative Matrix Factorization

8.3 Matrices Sharing Instances and Features

8.3.1 Unsupervised Learning: Clustering

8.3.2 Unsupervised Learning: Collaborative Matrix Factorization

8.4 Notes