Multi-aspect Learning: Methods and Applications (Intelligent Systems Reference Library, 242)

✍ Scribed by Richi Nayak, Khanh Luong

• Publisher: Springer
• Year: 2023
• Tongue: English
• Leaves: 191
• Edition: 1st ed. 2023
• Category: Library

✦ Synopsis

This book offers a detailed and comprehensive analysis of multi-aspect data learning, focusing especially on representation learning approaches for unsupervised machine learning. It covers state-of-the-art representation learning techniques for clustering and their applications in various domains. This is the first book to systematically review multi-aspect data learning, incorporating a range of concepts and applications. Additionally, it is the first to comprehensively investigate manifold learning for dimensionality reduction in multi-view data learning. The book presents the latest advances in matrix factorization, subspace clustering, spectral clustering and deep learning methods, with a particular emphasis on the challenges and characteristics of multi-aspect data. Each chapter includes a thorough discussion of state-of-the-art of multi-aspect data learning methods and important research gaps. The book provides readers with the necessary foundational knowledge to apply these methods to new domains and applications, as well as inspire new research in this emerging field.

✦ Table of Contents

Contents
1 Multi-aspect Data Learning: Overview, Challenges and Approaches
1.1 Introduction
1.2 Multi-type Relational Data
1.2.1 Object Type, Objects
1.2.2 Relationships
1.2.3 Bi-type Data
1.3 Multi-view Data
1.3.1 Partial Multi-view Data
1.3.2 Consensus and Complementary Information in Multi-view Data
1.4 Relationship Between MTRD and Multi-view Data
1.5 Real-World Multi-aspect Data Applications
1.5.1 Text Mining
1.5.2 Image Mining
1.5.3 Bio-informatics
1.5.4 Social-Network Mining
1.6 Multi-aspect Data Clustering
1.6.1 Why Design Customized Multi-aspect Clustering Methods?
1.6.2 Multi-aspect Data Clustering: Approaches
1.7 Chapter Conclusion
References
2 Non-negative Matrix Factorization-Based Multi-aspect Data Clustering
2.1 Introduction
2.2 NMF Framework: Introduction
2.3 NMF Framework: Basic Concepts and Definitions
2.3.1 NMF Formulation on Traditional Data
2.3.2 NMF-Based Clustering Process
2.3.3 Constraints
2.3.4 Benefits of NMF
2.4 NMF-Based Clustering Methods on One-Aspect Data
2.5 NMF-Based Clustering Methods on Multi-aspect Data
2.5.1 NMF-Based Clustering Methods on Multi-view Data
2.5.2 NMF-Based Clustering Methods on MTRD Data
2.6 Chapter Conclusion
References
3 NMF and Manifold Learning for Multi-aspect Data
3.1 Introduction
3.2 Introduction to Manifold Learning
3.3 NMF and Manifold Learning-Based Clustering Methods …
3.3.1 Manifold Learning Formulation
3.3.2 Manifold Learning Based NMF Methods of Traditional One-Aspect Data
3.3.3 Challenges in Manifold Learning
3.4 NMF and Manifold Learning-Based Clustering Methods …
3.4.1 Learning the Manifold on Each Aspect of the Multi-aspect Data
3.4.2 Learning the Accurate Manifold on Each Aspect
3.4.3 Learning the Intrinsic Consensus Manifold on Multi-view Data
3.4.4 Discussion: Manifold Learning Approaches for Multi-view Data
3.5 Chapter Conclusion
References
4 Subspace Learning for Multi-aspect Data
4.1 Introduction
4.2 Subspace Clustering: Basics
4.3 One-Aspect Subspace Clustering
4.3.1 Problem Definition
4.3.2 Subspace-Based Clustering Methods
4.4 Multi-aspect Subspace Clustering
4.4.1 Multi-view Clustering Definition
4.4.2 Multi-view Subspace Clustering Using Early Integration
4.4.3 Multi-view Subspace Clustering Using Late Integration
4.4.4 Multi-view Subspace Clustering Using Intermediate Integration
4.4.5 Multi-view Subspace Clustering Using a Shared Unified Representation
4.5 Discussion
4.6 Chapter Conclusion
References
5 Spectral Clustering on Multi-aspect Data
5.1 Introduction
5.2 Spectral Clustering on Traditional One-Aspect Data
5.2.1 Fundamental Concepts
5.2.2 Spectral Clustering Approach on One-Aspect Data
5.3 Spectral Clustering Methods on Multi-aspect Data
5.3.1 Multi-view Spectral Clustering Methods
5.3.2 MTRD Spectral Clustering
5.4 Chapter Conclusion
References
6 Learning Consensus and Complementary Information for Multi-aspect Data Clustering
6.1 Introduction
6.2 Overview of Learning Consensus and Complementary Information
6.3 Learning Consensus and Complementary Information Using NMF
6.3.1 NMF-Based Methods Focused on Learning the Consensus Information
6.3.2 NMF-Based Methods Focused on Enhancing the Complementary Information
6.3.3 NMF-Based Methods Focused on Enhancing Consensus and Complementary Information Both
6.4 Learning Consensus and Complementary Information Using Subspace
6.4.1 Subspace-Based Methods Learning the Consensus Information
6.4.2 Subspace-Based Methods Learning the Complementary Information
6.4.3 Subspace-Based Methods Learning Both Consensus and Complementary Information
6.5 Learning Consensus and Complementary Information Using Spectral Clustering
6.5.1 Spectral Methods Learning Consensus and Complementary Information
6.6 Summary of Constraints and Regularizations Designed …
6.6.1 For Learning the Consensus Information
6.6.2 For Learning the Complementary Information
6.7 Chapter Conclusion
References
7 Deep Learning-Based Methods for Multi-aspect Data Clustering
7.1 Introduction
7.2 Autoencoder-Based Multi-view Data Clustering
7.2.1 Introduction to Autoencoder
7.2.2 AE-Based Clustering for One-View Data
7.2.3 AE-Based Clustering for Multi-view Data
7.3 GAN-Based Multi-view Data Clustering
7.3.1 Introduction to GAN
7.3.2 GAN-Based Clustering for One-View Data
7.3.3 GAN-Based Clustering for Multi-view Data
7.4 Deep Matrix Factorization-Based Multi-view Clustering
7.4.1 Deep Matrix Factorization (DMF)-Based Framework Definition
7.4.2 DMF-Based Clustering on One-View Data
7.4.3 DMF-Based Clustering on Multi-view Data
7.4.4 Remarks: Deep Semi-NMF and Deep-NMF
7.5 Chapter Conclusion
References