Recommender Systems
Frontiers and Practices
- 2024
- Book
- Authors
- Dongsheng Li
- Jianxun Lian
- Le Zhang
- Kan Ren
- Tun Lu
- Tao Wu
- Xing Xie
- Publisher
- Springer Nature Singapore
About this book
This book starts from the classic recommendation algorithms, introduces readers to the basic principles and main concepts of the traditional algorithms, and analyzes their advantages and limitations. Then, it addresses the fundamentals of deep learning, focusing on the deep-learning-based technology used, and analyzes problems arising in the theory and practice of recommender systems, helping readers gain a deeper understanding of the cutting-edge technology used in these systems. Lastly, it shares practical experience with Microsoft 's open source project Microsoft Recommenders. Readers can learn the design principles of recommendation algorithms using the source code provided in this book, allowing them to quickly build accurate and efficient recommender systems from scratch.
Advertisement
Table of Contents
-
Frontmatter
-
Chapter 1. Overview of Recommender Systems
Dongsheng Li, Jianxun Lian, Le Zhang, Kan Ren, Tun Lu, Tao Wu, Xing XieThis chapter offers a detailed exploration of recommender systems, tracing their evolution from the 1980s to the present. It discusses the historical context that led to their development, the fundamental principles behind various recommendation algorithms, and the revolutionary impact of deep learning on the field. Additionally, the chapter delves into the diverse applications of recommender systems in industries such as e-commerce, content platforms, and marketing, highlighting their commercial value and technical similarities with search and advertising. The chapter also introduces the new paradigm of deep learning-based recommender systems, emphasizing the importance of representation learning and interaction function learning. Finally, it provides an overview of the technical architecture of recommender systems, differentiating between small- and large-scale systems, and concludes with a summary of the unique challenges and solutions in various industries.AI Generated
This summary of the content was generated with the help of AI.
AbstractThis chapter first introduces the history of the recommender system and the revolutionary changes in the field of recommender systems. Then, this chapter introduces the basic principles of recommender systems, including introducing the basic assumptions of recommendation algorithms from the perspective of machine learning, introducing how to define the recommendation problem in the form of a machine learning problem, and emphatically introducing the deep learning-based paradigm to solve the recommendation problem—“representation learning + interaction function learning”. This chapter also gives an overview of the technical architecture of recommender systems, including the differences between small- and medium-scale recommender systems and large-scale recommender systems. Finally, this chapter introduces the main application areas of recommender systems, such as e-commerce, content platforms, etc., and the actual business value brought by recommender systems to these application areas and compares the three main applications in the Internet field—search, advertising, and recommendation, by the differences and connections among them. Starting from industry problems, this chapter summarizes the differences in the application of recommender systems in different industries and outlines the solutions to different types of problems. -
Chapter 2. Classic Recommendation Algorithms
Dongsheng Li, Jianxun Lian, Le Zhang, Kan Ren, Tun Lu, Tao Wu, Xing XieThis chapter introduces classic recommendation algorithms that predate the rise of deep learning, focusing on content-based and collaborative filtering methods. It explains how to model structured and unstructured content for recommendations and delves into three mainstream collaborative filtering methods: memory-based, matrix factorization, and factorization machine methods. The chapter also highlights the advantages and limitations of content-based recommendations and discusses advanced memory-based collaborative filtering algorithms like SLIM and SSLIM. It concludes by summarizing the connections and differences between matrix factorization and factorization machines, providing a solid foundation for understanding the evolution of recommendation systems.AI Generated
This summary of the content was generated with the help of AI.
AbstractThis chapter introduces four types of classic recommendation algorithms, including content-based recommendation algorithms, classic collaborative filtering algorithms, matrix factorization methods, and factorization machines. Before the emergence of deep learning, these methods were the most mainstream techniques for recommender systems, widely recognized by both academia and industry. Although after the emergence of deep learning, these technologies are no longer the first choice of the industry, but the basic ideas and practical experience extracted from these technologies still affect the follow-up research. Therefore, in many deep learning-based recommendation algorithms, we can often see the reflections of the above approaches. -
Chapter 3. Foundations of Deep Learning
Dongsheng Li, Jianxun Lian, Le Zhang, Kan Ren, Tun Lu, Tao Wu, Xing XieThe chapter begins by explaining the fundamental structure and feedforward computation of neural networks, followed by an in-depth look at the back-propagation algorithm used for optimizing neural network models. It then delves into various types of deep neural networks, including convolutional neural networks (CNNs) which are essential for image analysis, recurrent neural networks (RNNs) designed for sequence data, and attention mechanisms that enhance the performance of models by focusing on relevant inputs. Additionally, the chapter explores sequence modeling techniques and pre-training methods such as Word2Vec, Transformer, and BERT, which have revolutionized the field of natural language processing. This comprehensive overview of deep learning concepts and models is designed to equip readers with the knowledge necessary to design and implement effective recommendation systems and other AI applications.AI Generated
This summary of the content was generated with the help of AI.
AbstractThis chapter introduces the basics of deep learning, including feedforward computation and backpropagation algorithms for deep neural networks, as well as various classic neural network models. As readers learn, they can combine the content of other chapters in this book to understand and design different types of neural network models for recommendation scenarios, taking into account the data characteristics and task properties, in order to improve recommendation performance. -
Chapter 4. Deep Learning-Based Recommendation Algorithms
Dongsheng Li, Jianxun Lian, Le Zhang, Kan Ren, Tun Lu, Tao Wu, Xing XieDeep learning technology has significantly transformed the progress of AI, particularly in the realm of recommendation systems. This chapter delves into the six most crucial topics in deep learning-based recommendation algorithms, including collaborative filtering, feature product, graph learning, sequential recommendation, knowledge distillation, and deep reinforcement learning. It examines how deep learning has enhanced the modeling and generalization capabilities of conventional recommendation algorithms and discusses novel algorithms inspired by deep learning. The chapter also explores the strong capabilities of deep learning in representation and generalization, making it a powerful tool for enhancing various recommendation algorithms. By understanding these topics, readers will gain valuable insights into the cutting-edge developments in recommendation systems and their practical applications across different industries.AI Generated
This summary of the content was generated with the help of AI.
AbstractThis chapter introduces the relationship between collaborative filtering and deep learning and then presented various deep learning-based collaborative filtering algorithms. Leveraging cutting-edge methods from deep learning, these algorithms can significantly improve the accuracy, scalability, diversity, and interpretability of recommendation systems, offering richer technological choices for recommendation system design. However, most of these algorithms are optimized for specific problems, and there are often limitations in practical applications. Therefore, at the system design level, algorithm integration or fusion needs to be considered. -
Chapter 5. Recommender System Frontier Topics
Dongsheng Li, Jianxun Lian, Le Zhang, Kan Ren, Tun Lu, Tao Wu, Xing XieThe chapter delves into the rapidly evolving field of recommender systems, highlighting three key research hotspots: conversational, causal, and common-sense recommendations. It also addresses the significant challenges faced by recommender systems, such as multi-source data fusion, scalability, and evaluation methods. Additionally, the chapter emphasizes the importance of responsible recommendation practices, including user privacy protection and explainability, to ensure the ethical and effective use of these systems. By exploring these topics, the chapter provides a comprehensive overview of the current state and future directions in recommender system research.AI Generated
This summary of the content was generated with the help of AI.
AbstractThis chapter introduces the hotspots of recommender system research, the key challenges of recommender system application, and how to achieve responsible recommendation technically. These contents may become the key of recommender system research and application in the future, so they need the continuous attention of researchers and developers. -
Chapter 6. Practical Recommender System
Dongsheng Li, Jianxun Lian, Le Zhang, Kan Ren, Tun Lu, Tao Wu, Xing XieThe chapter delves into the practical implementation of recommender systems, focusing on aspects such as data management, preprocessing, and algorithm selection. It introduces Microsoft Recommenders as a tool for best practices and provides hands-on code examples for various recommender models, including collaborative filtering, sequential models, and knowledge graph-based systems. The chapter also discusses the evaluation metrics and methods specific to recommender systems, ensuring a comprehensive guide for building and deploying industry-grade recommender systems.AI Generated
This summary of the content was generated with the help of AI.
AbstractThis chapter focuses on some problems and considerations in industry applications of recommender systems, and discusses the details of these actual applications based on the code in the Microsoft Recommenders repository and a cloud-based reference architecture. Readers are encouraged to follow the steps and methods in the text in a hands-on manner and experiment using the algorithms introduced in previous chapters. -
Chapter 7. Summary and Outlook
Dongsheng Li, Jianxun Lian, Le Zhang, Kan Ren, Tun Lu, Tao Wu, Xing XieThis chapter delves into the transformation of recommender systems, highlighting the shift from content-based and collaborative filtering algorithms to deep learning-based models. It discusses the advantages of representation learning and interaction function learning in capturing complex user-item relationships. Additionally, the chapter addresses the ethical considerations and challenges in recommender system development, such as privacy, bias, and social impact. It offers practical guidance through the Microsoft Recommenders open-source project, enabling readers to build efficient and responsible recommender systems.AI Generated
This summary of the content was generated with the help of AI.
AbstractThis chapter provides a summary of the book and offers insights into future trends in the research and application of recommender systems.
- Title
- Recommender Systems
- Authors
-
Dongsheng Li
Jianxun Lian
Le Zhang
Kan Ren
Tun Lu
Tao Wu
Xing Xie
- Copyright Year
- 2024
- Publisher
- Springer Nature Singapore
- Electronic ISBN
- 978-981-9989-64-5
- Print ISBN
- 978-981-9989-63-8
- DOI
- https://doi.org/10.1007/978-981-99-8964-5
Accessibility information for this book is coming soon. We're working to make it available as quickly as possible. Thank you for your patience.