Supervised Descriptive Pattern Mining - Luna, José María

. .

Résumé :
This book provides a general and comprehensible overview of supervised descriptive pattern mining, considering classic algorithms and those based on heuristics. It provides some formal definitions and a general idea about patterns, pattern mining, the usefulness of patterns in the knowledge discovery process, as well as a brief summary on the tasks related to supervised descriptive pattern mining. It also includes a detailed description on the tasks usually grouped under the term supervised descriptive pattern mining: subgroups discovery, contrast sets and emerging patterns. Additionally, this book includes two tasks, class association rules and exceptional models, that are also considered within this field. A major feature of this book is that it provides a general overview (formal definitions and algorithms) of all the tasks included under the term supervised descriptive pattern mining. It considers the analysis of different algorithms either based on heuristics or based on exhaustive search methodologies for any of these tasks. This book also illustrates how important these techniques are in different fields, a set of real-world applications are described. Last but not least, some related tasks are also considered and analyzed. The final aim of this book is to provide a general review of the supervised descriptive pattern mining field, describing its tasks, its algorithms, its applications, and related tasks (those that share some common features).This book targets developers, engineers and computer scientists aiming to apply classic and heuristic-based algorithms to solve different kinds of pattern mining problems and apply them to real issues. Students and researchers working in this field, can use this comprehensive book (which includes its methods and tools) as a secondary textbook.

Sommaire:

Chapter 1. Introduction to Pattern Mining

1.1 Importance of patterns

1.2 Type of patterns

1.3 Quality measures in pattern mining

1.3.1 Objective interestingness measures

1.3.2 Subjective interestingness measures

1.4 Scalability issues

1.4 Supervised descriptive local patterns

Chapter 2. Subgroup Discovery

2.1 Introduction

2.2 Task definition

2.3 Quality measures

2.4 Models in subgroup discovery

Chapter 3. Contrast sets

3.1 Introduction

3.2 Task definition

3.3 Algorithms

Chapter 4. Emerging patterns

4.1 Introduction

4.2 Task definition

4.3 Algorithms

Chapter 5. Class Association rules

5.2 Task definition

5.2.1 Association rules

5.2.2 Class association rules

5.2.3 Associative classification

5.3 Algorithms

Chapter 6. Exceptional models

6.1 Introduction

6.2 Exceptional model mining

6.3 Exceptional preference mining

6.4 Exceptional pattern mining

6.5 Algorithms

Chapter 7. Applications of supervised descriptive local patterns

7.1 Introduction

7.2 Subgroup discovery

7.3 Contrast sets

7.4 Emerging patterns

7.5 Exceptional models

7.6 Class association rules

Chapter 8. Additional tasks related to supervised pattern mining

8.1 Change mining

8.2 Mining of closed sets for labeled data

8.3 Bump hunting

8.4 Impact rules

8.5 Discrimination discovery

8.6 Context aware