دانلود کتاب Recommender Systems: Frontiers and Practices

فهرست مطالب :

Foreword
Recommender Systems: Navigators in the Ocean of Information
Preface
Contents
1 Overview of Recommender Systems
1.1 History of Recommender Systems
1.1.1 Content-Based Recommendation Algorithms
1.1.2 Collaborative Filtering-Based RecommendationAlgorithms
Key Steps in Collaborative Filtering Algorithms
Classification of Collaborative Filtering Algorithms
1.1.3 Deep Learning-Based Recommendation Algorithms
1.2 Principles of Recommender Systems
1.2.1 Recommender Systems from the Perspective of Machine Learning
Data Collection
Data Preprocessing
Recommendation Algorithm Selection and Model Training
Recommendation Performance Evaluation
Online Deployment and User Feedback
1.2.2 A New Paradigm for Deep Learning-Based Recommender System
User Representation Learning
Item Representation Learning
Interaction Function Learning
1.2.3 Common Architectures for Recommender Systems
Architecture of Small- and Medium-Scale Recommender Systems
Architecture of Large-Scale Recommender Systems
1.3 Values of Recommender Systems
1.3.1 Business Values of Recommender Systems
1.3.2 Recommendation, Search, and Advertising
Application Differences of the Three
Technical Similarities Among the Three
1.3.3 Industry Applications of Recommender Systems
E-Commerce Platforms
Content Platforms
Daily Services
Social Networks
Marketing and Sales
1.4 Summary
References
2 Classic Recommendation Algorithms
2.1 Content-Based Recommendation Algorithm
2.1.1 Recommendations Based on Structured Content
Basic Content-Based Recommendation Algorithms
Nearest Neighbor Classification Algorithm
Relevance Feedback-Based Algorithm
Decision Tree-Based Recommendation
Naive Bayes Classification
Linear Classification-Based Content Recommendation Algorithm
2.1.2 Recommendations Based on Unstructured Content
Text Representation
Representation of Non-text
2.1.3 Advantages and Limitations of Content-Based Recommendation
Advantages of Content-Based Recommendation
Limitations of Content-Based Recommendation
2.2 Collaborative Filtering-Based Recommendation Algorithms
2.2.1 Memory-Based Collaborative Filtering
Classic Memory-Based Collaborative Filtering Algorithms
Advanced Memory-Based Collaborative Filtering Algorithms
Summary of Memory-Based Collaborative Filtering Algorithms
2.2.2 Matrix Factorization Method and Factorization Machine Method
Matrix Factorization Method
Factorization Machine
The Connections and Differences Between Matrix Factorization and Factorization Machines
2.3 Summary
References
3 Foundations of Deep Learning
3.1 Neural Networks and Feedforward Computation
3.2 Back-Propagation Algorithm
3.3 Various Types of Deep Neural Networks
3.3.1 Convolutional Neural Network
3.3.2 Recurrent Neural Networks
3.3.3 Attention Mechanism
3.3.4 Sequence Modeling and Pre-training
Word2Vec
Transformer
BERT
3.4 Conclusion
References
4 Deep Learning-Based Recommendation Algorithms
4.1 Deep Learning and Collaborative Filtering
4.1.1 Restricted Boltzmann Machine-Based Collaborative Filtering
4.1.2 Autoencoder-Based Collaborative Filtering
4.1.3 Deep Learning and Matrix Factorization
Neural Collaborative Filtering
Deep Matrix Factorization
4.1.4 Neighborhood-Based Collaborative Filtering
4.2 Deep Learning and Feature Interaction
4.2.1 AFM Algorithm
4.2.2 PNN Algorithm
Inner Product-Based Interaction
Outer Product-Based Interaction
4.2.3 Wide and Deep Algorithm
4.2.4 DeepFM Algorithm
4.2.5 DCN Algorithm
4.2.6 xDeepFM Algorithm
4.2.7 AutoInt Algorithm
4.2.8 Additional Thoughts on Feature Interaction
4.3 Graph Representation Learning and Recommender System
4.3.1 Graph Embedding and Fundamentals of Graph Neural Network
Graph Embedding
Graph Neural Network
4.3.2 Graph Neural Network and Collaborative Filtering
GCMC Algorithm
NGCF Algorithm
LightGCN Algorithm
4.3.3 Graph Neural Network and Social Recommendation
GraphRec Algorithm
DiffNet Algorithm
4.4 Sequential Recommender Systems
4.4.1 Motivation, Definition, and Classification of Sequential Recommendation
4.4.2 Classification of Sequential Recommendation Algorithms
Sequential Pattern Mining
Latent Factor Representation
Markov Chain-Based Sequential Recommendation
Deep Learning-Based Sequential Recommendation
4.4.3 Recurrent Neural Network-Based Sequential Recommendation
4.4.4 Non-autoregressive Neural Network-Based Sequence Modeling
4.4.5 Self-attention-Based Sequence Recommendation
4.4.6 Memory-Based Neural Networks for Sequential Recommendation
4.4.7 User–Item Dual Sequence Modeling
4.5 Recommender Systems Combined with Knowledge Graph
4.5.1 Enhancing User–Item Interaction Modeling
The RippleNet Model
The KGAT Model
4.5.2 Joint Learning of Graph Modeling and Item Recommendation
The KTUP Model
The MKR Model
4.5.3 Knowledge Graph Enhanced Item Representation
The DKN Model
The KRED Model
4.5.4 Explainability
The KPRN Model
The PGPR Model
The ADAC Model
4.6 Reinforcement Learning-Based Recommendation Algorithms
4.6.1 Multi-armed Bandit-Based Recommendation Algorithms
4.6.2 Introduction to Reinforcement Learning
4.6.3 Reinforcement Learning-Based Recommendation Algorithms
4.6.4 Modeling and Optimization of Deep Reinforcement Learning
4.7 Conclusion
References
5 Recommender System Frontier Topics
5.1 Recommendation Algorithm Hotspots
5.1.1 Conversational Recommenders
Exploration–Exploitation Trade-Off in Cold-Start Scenarios
Question-Centered Conversational Recommendation
Strategy-Centered Conversational Recommendation
Dialogue Understanding and Generation
5.1.2 Causal Recommendation
5.1.3 Common-Sense Recommendation
5.2 Application Challenges for Recommender Systems
5.2.1 Multi-source Data Fusion
5.2.2 Scalability
Clustering
Dimension Reduction
Distributed Computing
Incremental Recommendation
5.2.3 Performance Evaluation
Offline Evaluation
Online Evaluation
User Study
5.2.4 Cold-Start Problem
Popular Item Recommendation
Recommendation with Additional Information
Recommendation with Expert Annotations
Conversational Recommendation
5.3 Responsible Recommendation
5.3.1 User Privacy
5.3.2 Explainability
Technical Routes to Recommendation Generation
Presentation of Explainable Recommendation
5.3.3 Algorithm Bias
5.4 Summary
References
6 Practical Recommender System
6.1 Introduction
6.2 Architecture and Implementation of Industry-Grade Recommender System
6.2.1 Characteristics of Industry-Grade Recommender System
6.2.2 Commonly Used Architecture of Recommender System
6.2.3 Offline Recommender System
Real-Time Recommender System
6.2.4 Industrial Implementation of Recommender System
6.3 Practices of Recommender System
6.3.1 Data Management and Preprocessing
Data Split
Data Transformation
6.3.2 Algorithm Selection and Model Training
Spark-Based ALS
The Implementation of the Sequential Recommender Model
Knowledge Graph-Based Recommender System
6.3.3 Evaluation Metrics and Methods
6.4 Development and Operation of Recommender Systems in the Cloud
6.4.1 Advantages of Using Cloud for Recommender Systems
6.4.2 Cloud-Based Development and Operations of Recommender Systems
6.5 Summary
References
7 Summary and Outlook