Intro
General Chairs' Preface
Program Chairs' Preface
Organization
Contents - Part II
Recommendation Systems
MDKE: Multi-level Disentangled Knowledge-Based Embedding for Recommender Systems
1 Introduction
2 Preliminaries
3 Methodology
3.1 Item Content Extraction
3.2 User Preference Propagation
3.3 Graph Structural Embedding
3.4 Model Prediction and Training
3.5 Analysis of MDKE
4 Experiments
4.1 Experimental Settings
4.2 MDKE Performance
4.3 Impacts of Hyperparameters
4.4 Ablation Study
5 Related Work

5.1 Graph-Based Methods for Recommendation
5.2 Disentangled Representation Learning
6 Conclusion
References
M3-IB: A Memory-Augment Multi-modal Information Bottleneck Model for Next-Item Recommendation
1 Introduction
2 Related Work
3 Problem Definition
4 The Proposed Method
4.1 Memory Network Framework
4.2 Multi-modal Information Bottleneck Model
4.3 Implementation for Next-Item Recommendation
4.4 Model Learning and Complexity Analysis
5 Experiment
5.1 Experimental Setup
5.2 Model Comparison
5.3 Ablation Study
5.4 Hyper-Parameter Study

6 Conclusion
References
Fully Utilizing Neighbors for Session-Based Recommendation with Graph Neural Networks
1 Introduction
2 Related Work
3 Preliminaries
3.1 Problem Definition
3.2 Session Graph Construction
3.3 Graph Attention Diffusion
4 Methodology
4.1 Positional Graph Attention Aggregation Layer
4.2 Multi-head Graph Attention Diffusion Layer
4.3 Session Embedding Readout
4.4 Prediction and Training
5 Experiments
5.1 Datasets
5.2 Baselines and Evaluation Metrics
5.3 Implementation Details
5.4 Overall Comparison (RQ1)
5.5 Ablation Study

5.6 Hyper-parameter Analysis (RQ4)
6 Conclusion
References
Inter- and Intra-Domain Relation-Aware Heterogeneous Graph Convolutional Networks for Cross-Domain Recommendation
1 Introduction
2 Related Work
2.1 Graph-Based Recommendation
2.2 Cross-Domain Recommendation
3 Preliminaries
4 Proposed Framework
4.1 Graph Construction and Embedding
4.2 Relation-Aware GCN Layer
4.3 Gating Fusion Layer
4.4 Prediction Layer
4.5 Model Training
5 Experiments
5.1 Experimental Settings
5.2 RQ1: Performance Comparison
5.3 RQ2: Ablation Study

5.4 RQ3: Effect of Inter-domain Relations
5.5 RQ4: Parameter Analysis
6 Conclusion and Future Work
References
Enhancing Graph Convolution Network for Novel Recommendation
1 Introduction
2 Related Work
2.1 Graph Based Methods for Recommendation
2.2 Novel Recommendation
3 Our Proposed Model
3.1 Problem Formulation
3.2 Model Overview
3.3 Embedding Layer
3.4 Masking Layer
3.5 Graph Convolutional Layer
3.6 Negative Sampling Layer
3.7 Reconstruction Layer
3.8 Gated Fusion Layer
3.9 Training
4 Experiments
4.1 Experimental Settings