Knowledge-aware Graph Neural Networks with Label Smoothness Regularization for Recommender Systems

By Hongwei Wang, Fuzheng Zhang, Mengdi Zhang, Jure Leskovec, Miao Zhao, Wenjie Li, Zhongyuan Wang. Knowledge Discovery and Data Mining, 2019

660 citations recorded by the source corpus.

Knowledge graphs capture structured information and relations between a set of entities or items. As such knowledge graphs represent an attractive source of information that could help improve recommender systems. However, existing approaches in this domain rely on manual feature engineering and do not allow for an end-to-end training. Here we propose Knowledge-aware Graph Neural Networks with Label Smoothness regularization (KGNN-LS) to provide better recommendations. Conceptually, our approach computes user-specific item embeddings by first applying a trainable function that identifies important knowledge graph relationships for a given user. This way we transform the knowledge graph into a user-specific weighted graph and then apply a graph neural network to compute personalized item embeddings. To provide better inductive bias, we rely on label smoothness assumption, which posits that adjacent items in the knowledge graph are likely to have similar user relevance labels/scores. Label smoothness provides regularization over the edge weights and we prove that it is equivalent to a label propagation scheme on a graph. We also develop an efficient implementation that shows strong scalability with respect to the knowledge graph size. Experiments on four datasets show that our method outperforms state of the art baselines. KGNN-LS also achieves strong performance in cold-start scenarios where user-item interactions are sparse.

Subfields: Artificial Intelligence, Information Systems

View Paper: https://www.semanticscholar.org/paper/594dc362b4332ae661e3d71da17d097bb4a357dd

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Xiangnan He
2. Tat-seng Chua
3. Xiang Wang
4. Xing Xie
5. Meng Wang
6. Jie Tang
7. Le Wu
8. Fangzhao Wu

In Information Systems

1. Xiangnan He
2. Xiang Wang
3. Tat-seng Chua
4. Xing Xie
5. Meng Wang
6. Le Wu
7. Kun Zhang
8. Fangzhao Wu

Citations in the corpus

• Are we really making much progress?: Revisiting, benchmarking and refining heterogeneous graph neural networks (2021, background): “We also tried to stack more graph layers, but performance deteriorates when we do that, which is also found in [33, 34] experiments.”
• Multi-modal Knowledge Graphs for Recommender Systems (2020, methodology): “Given the sub-KG, [26] collect Satori IDs of all valid movies by matching their names with tail of triples.”
• Graph Random Neural Networks for Semi-Supervised Learning on Graphs (2020, background): “igm of graph mining from structural explorations to representation learning. A wide variety of graph applications have benefited from this shift, such as node classification [10, 27], link prediction [45, 48, 50], and graph classification [32, 46]. The essential procedure in GNNs is the feature prop”
• Graph Neural Networks in Recommender Systems: A Survey (2020, methodology): “MovieLens-20M KGCN [147], KGNN-LS [145], CKAN [162]”
• A Survey on Knowledge Graph-Based Recommender Systems (2020, methodology): “[2], [14], [44], [45], [66], [70], [81], [83], [87], [92], [93], [95], [96] MovieLens-20M [44], [86], [88], [89], [91] museums, parks, cities, etc.) to users based on their historical check-in data.”
• WikiGraphs: A Wikipedia Text - Knowledge Graph Paired Dataset (2021, methodology): “Hand labelling text articles with KGs is expensive and not scalable (Lebret et al., 2016), therefore we utilize an existing and established knowledge base, Freebase (Bollacker et al., 2008), and designed an automated process to extract a relevant subgraph from it for each Wikipedia article.”
• Learning Intents behind Interactions with Knowledge Graph for Recommendation (2021, methodology): “Experimental results on three benchmark datasets show that, KGIN achieves significant improvements over the state-ofthe-art methods like KGAT [41], KGNN-LS [38], and CKAN [47].”
• A federated graph neural network framework for privacy-preserving personalization (2021, methodology): “Besides the user-item graph, several GNN-based recommendation methods also incorporate other kinds of graphs into recommendation, such as user-item-entity graph [29] and user-user-item graph [5].”
• A Survey of Knowledge-enhanced Text Generation (2020, background): “entities are interconnected with certain knowledge. Recent advances in artificial intelligence research have demonstrated the effectiveness of KG in various applications such as recommendation system [143]. †For brevity, “entities” denotes both entities (e.g., Prince) and concepts (e.g., musician) t”
• A Survey on Accuracy-Oriented Neural Recommendation: From Collaborative Filtering to Information-Rich Recommendation (2021, background): “[178], [179], and GNN-based approaches [97], [128], [141], [170], [180], [181], [182], [183], [184], [185], [186], [187].”
• RecBole: Towards a Unified, Comprehensive and Efficient Framework for Recommendation Algorithms (2020, neutral): “KGNN-LS [66] SIGKDD 2019 MovieLens, Book-Crossing, LastFM, Dianping-Food”