A Cardinality-Decomposed Loss (CDL) is proposed that combines both Cross Entropy (CE) and BPR to enable the model to collectively optimize for relations across cardinalities and is found that CDL consistently improves discriminability in attribute embeddings.
Abstract
Graph Neural Networks trained on heterogenous bipartite graphs form a common basis in recommendation systems. These graphs often express relations that vary in cardinality, for example, user-item preferences are one-to-many and user-attribute features are one-to-one. Traditionally, a unique loss function is applied for all of the network components which is often Bayesian Personalized Ranking (BPR). While BPR works well for the recommendation task, we find that it causes attribute embeddings to collapse to near-random geometry -- a silent failure that leaves standard ranking metrics largely unaffected and therefore invisible to conventional evaluation. This in turn pollutes user node embeddings, which are shaped by both edge types simultaneously, hurting downstream tasks like personalization, segmentation, etc. Here we propose a Cardinality-Decomposed Loss (CDL) that combines both Cross Entropy (CE) and BPR to enable the model to collectively optimize for relations across cardinalities. We confirm this CE-BPR conflict by showing the two losses compete in the shared encoder's parameter space. We evaluate CDL on five datasets spanning two structural configurations -- one-to-one attributes on user nodes (MovieLens-1M, Last.fm-360K, PayPal Audience Factory, BookCrossing) and on item nodes (Yelp) -- and find that CDL consistently improves discriminability in attribute embeddings. We also show that ranking (NDCG) improves when attributes carry meaningful preference signal, but conflicts with it when the correlation is weak. We use a lambda parameter to navigate this trade-off, and a lambda-sweep reveals that dataset behavior is governed by two graph properties -- semantic alignment and topology leakage. Semantic alignment measures whether the attribute predicts preferences, while topology leakage measures whether the graph's connectivity already encodes it.
HGAT-Rec is proposed, which incorporates a heterogeneity-aware contrastive learning (HCL) objective that grounds view construction and sample selection in the typed relational structure of a cross-domain heterogeneous graph: type-stratified edge dropout preserves high-signal interaction channels proportionally to their attention weight.
Rui Zhang, YaTing Zhao, FengBo Wang et al.· Journal of King Saud Univers...· 0 citations
This study proposes a wine recommendation system based on heterogeneous graph transformers that integrates hybrid node features from users, reviews, and wine knowledge. We use real-world data from a wine e-commerce platform, including 1685 wines, 12,361 users, 71,507 reviews, and 71,507 ratings. A multi-relational knowledge graph is constructed to represent relationships among wines, customers, and content. Node features are generated via decoding-enhanced BERT with disentangled attention–based review embeddings and structured wine attributes, while the graph structure captures both objective and subjective signals. We evaluate three heterogeneous graph transformer variants and benchmark them against the baseline light graph convolution network model. Our best-performing configuration, heterogeneous graph transformer model C with hybrid features, achieves an average improvement of 2.11% in precision and 5.87% in normalized discounted cumulative gain (NDCG) across top-K recommendations. These results highlight the value of feature diversity and graph-based modeling for enhancing personalized product recommendations in e-commerce.
J. Liou, Zi-Yi Hong, L. Cheng· Journal of information scien...· 0 citations
Orthogonal Decomposition for Social Recommendation (ODSR) is proposed, an embedding-space framework that orthogonally decomposes the aggregated social message into an aligned component and an orthogonal deviation, and learns a dimension-wise vector gate to regulate the deviation under ranking supervision.
Rongfeng Guo, Yinxuan Huang, Wei Chen et al.· Proceedings of the 32nd ACM...· 0 citations
A novel neural network called the Co-occurrence Graph Neural Network (CoGNN), which utilizes two co-occurrence graphs to establish user and item relationships and outperforms various baseline models in terms of recommendation accuracy and algorithm convergence.
The proposed LLM-augmented Multi-Graph Contrastive Learning (LLM-MGCL) is a multi-graph neural network that uses semantic and spatial information about items to extend the LightGCN backbone with two auxiliary item-item graphs that outperforms classical collaborative filtering, matrix factorization, and interaction-only graph neural network baselines.
Recommender systems increasingly incorporate graph embeddings and graph neural networks to capture high-order relationships between users and items. However, the additional complexity of these approaches does not necessarily guarantee better recommendation quality than strong classical and latent-factor baselines. This study presents a reproducible comparison of six recommendation models representing four methodological families: Logistic Regression and Random Forest; Matrix Factorization with Bayesian Personalized Ranking; DeepWalk and node2vec; and LightGCN. The experiments were conducted on the MovieLens 1M dataset using a per-user temporal split. For each user, the most recent positive interaction was assigned to testing, the preceding interaction to validation, and all earlier positive interactions to training. The primary evaluation used identical candidate sets containing one held-out positive movie and 99 sampled unobserved movies. Performance was measured using Recall, Precision, Hit Rate, and NDCG at multiple cutoffs, complemented by bootstrap confidence intervals, paired statistical tests, computational-efficiency measurements, and analyses by user activity and movie popularity. Matrix Factorization achieved the best overall performance, reaching a Recall@10 of 0.7458 and an NDCG@10 of 0.4558, representing an approximately 56% improvement in NDCG@10 over Random Forest, the strongest classical baseline. Validation-based tuning improved LightGCN to an NDCG@10 of 0.2875; it significantly outperformed Logistic Regression but remained statistically indistinguishable from Random Forest after Holm correction. Tuned node2vec also significantly outperformed DeepWalk, reaching an NDCG@10 of 0.1593, although both random-walk embedding methods’ results remained substantially below than the strongest baselines. Popularity-based analysis further revealed that classical models and LightGCN achieved substantially higher ranking effectiveness for popular movies, whereas Matrix Factorization maintained comparatively stronger performance for less-popular items. These findings show that under the evaluated setting, greater model complexity did not consistently translate into higher recommendation effectiveness, and they thus highlight the importance of strong baselines, model tuning, standardized evaluation, and reproducible experimental protocols.
R. Bojorque, David Yánez-Peter, Miguel Arcos-Argudo· Algorithms· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
