DGCN: Diversified Recommendation with Graph Convolutional Networks

Abstract

These years much effort has been devoted to improving the accuracy or relevance of the recommendation system. Diversity, a crucial factor which measures the dissimilarity among the recommended items, received rather little scrutiny. Directly related to user satisfaction, diversification is usually taken into consideration after generating the candidate items. However, this decoupled design of diversification and candidate generation makes the whole system suboptimal. In this paper, we aim at pushing the diversification to the upstream candidate generation stage, with the help of Graph Convolutional Networks (GCN). Although GCN based recommendation algorithms have shown great power in modeling complex collaborative filtering effect to improve the accuracy of recommendation, how diversity changes is ignored in those advanced works. We propose to perform rebalanced neighbor discovering, category-boosted negative sampling and adversarial learning on top of GCN. We conduct extensive experiments on real-world datasets. Experimental results verify the effectiveness of our proposed method on diversification. Further ablation studies validate that our proposed method significantly alleviates the accuracy-diversity dilemma.

WWW'21

References in corpus (28)

• Adam: A Method for Stochastic Optimization
• Distributed Representations of Words and Phrases and their Compositionality
• Semi-Supervised Classification with Graph Convolutional Networks
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Inductive Representation Learning on Large Graphs
• BPR: Bayesian Personalized Ranking from Implicit Feedback
• Neural Graph Collaborative Filtering
• Graph Convolutional Neural Networks for Web-Scale Recommender Systems
• Unsupervised Domain Adaptation by Backpropagation
• KGAT: Knowledge Graph Attention Network for Recommendation
• On the Convergence of Adam and Beyond
• Session-based Recommendation with Graph Neural Networks
• Simplifying Graph Convolutional Networks
• Graph Convolutional Matrix Completion
• xDeepFM: Combining Explicit and Implicit Feature Interactions for Recommender Systems
• Knowledge Graph Convolutional Networks for Recommender Systems
• Solving the apparent diversity-accuracy dilemma of recommender systems
• DeepFM: A Factorization-Machine based Neural Network for CTR Prediction
• Deep Learning Recommendation Model for Personalization and Recommendation Systems
• How Powerful are Graph Neural Networks?
• NPA: Neural News Recommendation with Personalized Attention
• Learning Tree-based Deep Model for Recommender Systems
• Fast Greedy MAP Inference for Determinantal Point Process to Improve Recommendation Diversity
• Expectation-Maximization for Learning Determinantal Point Processes
• Joint Optimization of Tree-based Index and Deep Model for Recommender Systems
• Pixie: A System for Recommending 3+ Billion Items to 200+ Million Users in Real-Time
• Recent Advances in Diversified Recommendation
• DPPy: Sampling DPPs with Python

Cited by in corpus (18)

• DGRec: Graph Neural Network for Recommendation with Diversified Embedding Generation
• Alleviating Matthew Effect of Offline Reinforcement Learning in Interactive Recommendation
• User-controllable Recommendation Against Filter Bubbles
• Feature-aware Diversified Re-ranking with Disentangled Representations for Relevant Recommendation
• DVR: Micro-Video Recommendation Optimizing Watch-Time-Gain under Duration Bias
• Contextual Distillation Model for Diversified Recommendation
• KHAN: Knowledge-Aware Hierarchical Attention Networks for Accurate Political Stance Prediction
• Are We Really Achieving Better Beyond-Accuracy Performance in Next Basket Recommendation?
• Uncovering the Deep Filter Bubble: Narrow Exposure in Short-Video Recommendation
• Practitioners Versus Users: A Value-Sensitive Evaluation of Current Industrial Recommender System Design
• Advancing Biomedicine with Graph Representation Learning: Recent Progress, Challenges, and Future Directions
• ROLeR: Effective Reward Shaping in Offline Reinforcement Learning for Recommender Systems
• DACSR: Decoupled-Aggregated End-to-End Calibrated Sequential Recommendation
• Topic-Level Bayesian Surprise and Serendipity for Recommender Systems
• HGCH: A Hyperbolic Graph Convolution Network Model for Heterogeneous Collaborative Graph Recommendation
• Comprehensive List Generation for Multi-Generator Reranking
• Contrastive Adaptive Propagation Graph Neural Networks for Efficient Graph Learning
• Category-based and Popularity-guided Video Game Recommendation: A Balance-oriented Framework