Augmenting the User-Item Graph with Textual Similarity Models
Not on the map. Why?
Abstract (arXiv)
This paper introduces a simple and effective form of data augmentation for recommender systems. A paraphrase similarity model is applied to widely available textual data, such as reviews and product descriptions, yielding new semantic relations that are added to the user-item graph. This increases the density of the graph without needing further labeled data. The data augmentation is evaluated on a variety of recommendation algorithms, using Euclidean, hyperbolic, and complex spaces, and over three categories of Amazon product reviews with differing characteristics. Results show that the data augmentation technique provides significant improvements to all types of models, with the most pronounced gains for knowledge graph-based recommenders, particularly in cold-start settings, leading to state-of-the-art performance.
Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 55 of its 97 references that are in the map. First posted on arXiv 20 Sep 2021.
Builds on
55 of 97 references.
References · 55
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
- 2014Adam: A Method for Stochastic Optimizationcited by 36,960
- 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,049
- 2014Convolutional Neural Networks for Sentence Classificationcited by 1,124
- 2017Neural Collaborative Filteringcited by 1,108
- 2016Geometric deep learning: going beyond Euclidean datacited by 1,042
- 2014Embedding Entities and Relations for Learning and Inference in Knowledge Basescited by 929
- 2016Complex Embeddings for Simple Link Predictioncited by 897
- 2019RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Spacecited by 893
- 2017An Overview of Multi-Task Learning in Deep Neural Networkscited by 791
- 2017Poincar'e Embeddings for Learning Hierarchical Representationscited by 674
- 2019Hyperbolic Graph Convolutional Neural Networkscited by 400
38 more
- 2018Hyperbolic Neural Networkscited by 393
- 2019KGAT: Knowledge Graph Attention Network for Recommendationcited by 342
- 2018Representation Tradeoffs for Hyperbolic Embeddingscited by 204
- 2020Low-Dimensional Hyperbolic Knowledge Graph Embeddingscited by 195
- 2019Quaternion Knowledge Graph Embeddingscited by 175
- 2020A Survey on Knowledge Graph-Based Recommender Systemscited by 170
- 2018Poincar'e GloVe: Hyperbolic Word Embeddingscited by 155
- 2018Canonical Tensor Decomposition for Knowledge Base Completioncited by 150
- 2018DKN: Deep Knowledge-Aware Network for News Recommendationcited by 137
- 2017Deep Learning based Recommender System: A Survey and New Perspectivescited by 136
- 2019Multi-relational Poincar'e Graph Embeddingscited by 126
- 2018Hyperbolic Attention Networkscited by 122
- 2019Reinforcement Knowledge Graph Reasoning for Explainable Recommendationcited by 102
- 2015Inferring Networks of Substitutable and Complementary Productscited by 91
- 2019Unifying Knowledge Graph Learning and Recommendation: Towards a Better Understanding of User Preferencescited by 87
- 2016Improved Recurrent Neural Networks for Session-based Recommendationscited by 86
- 2017Joint Deep Modeling of Users and Items Using Reviews for Recommendationcited by 84
- 2018Embedding Text in Hyperbolic Spacescited by 76
- 2017Translation-based Recommendationcited by 70
- 2019Multi-Task Feature Learning for Knowledge Graph Enhanced Recommendationcited by 70
- 2018Data augmentation instead of explicit regularizationcited by 62
- 2019Inferring Concept Hierarchies from Text Corpora via Hyperbolic Embeddingscited by 49
- 2018HyperML: A Boosting Metric Learning Approach in Hyperbolic Space for Recommender Systemscited by 47
- 2018Large-Margin Classification in Hyperbolic Spacecited by 41
- 2018Learning Heterogeneous Knowledge Base Embeddings for Explainable Recommendationcited by 39
- 2017Latent Relational Metric Learning via Memory-based Attention for Collaborative Rankingcited by 38
- 2020Computationally Tractable Riemannian Manifolds for Graph Embeddingscited by 29
- 2020How Useful are Reviews for Recommendation? A Critical Review and Potential Improvementscited by 24
- 2017Hyperbolic Representation Learning for Fast and Efficient Neural Question Answeringcited by 23
- 2019Relational Collaborative Filtering:Modeling Multiple Item Relations for Recommendationcited by 23
- 2019Every child should have parents: a taxonomy refinement algorithm based on hyperbolic term embeddingscited by 23
- 2019Research Commentary on Recommendations with Side Information: A Survey and Research Directionscited by 23
- 2020A Fully Hyperbolic Neural Model for Hierarchical Multi-Class Classificationcited by 23
- 2021Symmetric Spaces for Graph Embeddings: A Finsler-Riemannian Approachcited by 23
- 2015Ricci Curvature of the Internet Topologycited by 22
- 2019HyperKG: Hyperbolic Knowledge Graph Embeddings for Knowledge Base Completioncited by 20
- 2018Skip-gram word embeddings in hyperbolic spacecited by 19
- 2019Scalable Hyperbolic Recommender Systemscited by 19
50 most cited of 55 All 55 (JSON)
Built on by
Citing papers per year: 2021: 1; 2022: 0; 2023: 2.
3 papers.