| Description |
1 online resource (241 pages) |
| Contents |
Preface; Contents; 1 Introduction; 1.1 The Knowledge Discovery Process; 1.2 Preprocessing; 1.2.1 Data Preparation; 1.2.2 Data Reduction; 1.3 Data Mining; 1.3.1 Supervised Learning; 1.3.2 Unsupervised Learning; 1.3.3 Semi-supervised Learning; 1.3.4 Scalability Consideration; 1.4 Classification; 1.4.1 Validation Schemes; 1.4.2 Evaluation Measures; References; 2 Multiple Instance Learning; 2.1 Formal Description; 2.2 Origin of MIL; 2.2.1 Relationship with Propositional Learning; 2.2.2 Relationship with Relational Learning ; 2.3 MIL Paradigms; 2.3.1 Multi-instance Classification and Regression |
|
2.3.2 Multi-instance Clustering2.3.3 Instance Annotation; 2.4 Applications of MIL; 2.4.1 Bioinformatics; 2.4.2 Image Classification and Retrieval; 2.4.3 Web Mining and Text Classification; 2.4.4 Object Detection and Tracking; 2.4.5 Medical Diagnosis and Imaging; 2.4.6 Other Classification Applications; 2.4.7 Regression Applications; 2.4.8 Clustering Applications; References; 3 Multi-instance Classification; 3.1 Introduction; 3.2 Formal Description; 3.3 Taxonomy ; 3.4 MI Assumptions ; 3.4.1 Standard MI Assumption; 3.4.2 Weidmann et al.'s Hierarchy; 3.4.3 Collective Assumption |
|
3.4.4 Mixture Distribution Assumption3.4.5 Soft Bag MI Assumption; 3.5 Distance Metrics ; 3.5.1 Bags as Point Sets; 3.5.2 Bags as Probability Distributions; 3.6 Real-World Applications ; 3.6.1 Bioinformatics; 3.6.2 Image Classification and Retrieval; 3.6.3 Web Mining and Text Classification; 3.6.4 Medical Diagnosis and Imaging; 3.6.5 Acoustic Classification; 3.7 Some Comments on Software Tools ; References; 4 Instance-Based Classification Methods; 4.1 Introduction ; 4.2 Wrapper Methods to Single-Instance Learning Algorithms; 4.3 Maximum Likelihood-Based Methods |
|
4.3.1 Maximum Likelihood Principle4.3.2 Diverse Density; 4.3.3 Logistic Regression; 4.3.4 Boosting; 4.4 Decision Rules and Tree-Based Methods ; 4.5 Instance-Level SVM; 4.6 Neural Network-Based Methods ; 4.6.1 Feedforward Neural Networks; 4.6.2 Recurrent Neural Networks; 4.6.3 Decision-Based Neural Networks; 4.6.4 Network Combinations; 4.7 Evolutionary Based Methods ; 4.8 Experimental Analysis ; 4.8.1 Setup; 4.8.2 Results and Discussion; 4.9 Summarizing Comments; References; 5 Bag-Based Classification Methods; 5.1 Introduction; 5.2 Original Bag Space Methods; 5.2.1 Nearest Neighbor Methods |
|
5.2.2 Bag-Level SVM5.3 Mapped Bag Space Methods; 5.3.1 Mapping Methods Based on Bag Statistics; 5.3.2 Mapping Methods Based on Prototype Concatenation ; 5.3.3 Mapping Methods Based on Counting; 5.3.4 Mapping Methods Based on Distance; 5.3.5 Bag-Level Distance Mapping Methods; 5.4 Experimental Analysis ; 5.4.1 Setup; 5.4.2 Results and Discussion; 5.5 Comparing Instance-Based, Bag-Based, and Traditional Classification Methods; 5.6 Summarizing Comments; References; 6 Multi-instance Regression; 6.1 Introduction ; 6.2 MIR Formulation ; 6.2.1 Problem Description ; 6.2.2 Evaluation Measures |
| Summary |
This book provides a general overview of multiple instance learning (MIL), defining the framework and covering the central paradigms. The authors discuss the most important algorithms for MIL such as classification, regression and clustering. With a focus on classification, a taxonomy is set and the most relevant proposals are specified. Efficient algorithms are developed to discover relevant information when working with uncertainty. Key representative applications are included. This book carries out a study of the key related fields of distance metrics and alternative hypothesis. Chapters examine new and developing aspects of MIL such as data reduction for multi-instance problems and imbalanced MIL data. Class imbalance for multi-instance problems is defined at the bag level, a type of representation that utilizes ambiguity due to the fact that bag labels are available, but the labels of the individual instances are not defined. Additionally, multiple instance multiple label learning is explored. This learning framework introduces flexibility and ambiguity in the object representation providing a natural formulation for representing complicated objects. Thus, an object is represented by a bag of instances and is allowed to have associated multiple class labels simultaneously. This book is suitable for developers and engineers working to apply MIL techniques to solve a variety of real-world problems. It is also useful for researchers or students seeking a thorough overview of MIL literature, methods, and tools |
| Notes |
6.3 Instance-Based Regression Methods |
| Bibliography |
Includes bibliographical references |
| Notes |
Print version record |
| In |
Springer eBooks |
| Subject |
Machine learning.
|
|
Artificial intelligence.
|
|
Artificial Intelligence
|
|
Software
|
|
Machine Learning
|
|
artificial intelligence.
|
|
software.
|
|
Image processing.
|
|
Algorithms & data structures.
|
|
Artificial intelligence.
|
|
COMPUTERS -- General.
|
|
Machine learning
|
| Form |
Electronic book
|
| Author |
Herrera, Francisco
|
|
Ventura, S. (Sebastian)
|
|
Bello, Rafael
|
|
Cornelis, Chris
|
|
Zafra, Amelia
|
|
Sánchez-Tarragó, Dánel
|
|
Vluymans, Sarah
|
| ISBN |
9783319477596 |
|
3319477595 |
|
