DisCo: Graph-Based Disentangled Contrastive Learning for Cold-Start Cross-Domain Recommendation

• URL: http://arxiv.org/abs/2412.15005v3
• Date: Wed, 12 Feb 2025 04:50:07 GMT
• Title: DisCo: Graph-Based Disentangled Contrastive Learning for Cold-Start Cross-Domain Recommendation
• Authors: Hourun Li, Yifan Wang, Zhiping Xiao, Jia Yang, Changling Zhou, Ming Zhang, Wei Ju,
• Abstract summary: Cross-domain recommendation (CDR) has emerged as a promising solution. However, users with similar preferences in the source domain may exhibit different interests in the target domain. We propose a novel graph-based disentangled contrastive learning framework to capture fine-grained user intent.
• Score: 11.61586672399166
• License:
• Abstract: Recommender systems are widely used in various real-world applications, but they often encounter the persistent challenge of the user cold-start problem. Cross-domain recommendation (CDR), which leverages user interactions from one domain to improve prediction performance in another, has emerged as a promising solution. However, users with similar preferences in the source domain may exhibit different interests in the target domain. Therefore, directly transferring embeddings may introduce irrelevant source-domain collaborative information. In this paper, we propose a novel graph-based disentangled contrastive learning framework to capture fine-grained user intent and filter out irrelevant collaborative information, thereby avoiding negative transfer. Specifically, for each domain, we use a multi-channel graph encoder to capture diverse user intents. We then construct the affinity graph in the embedding space and perform multi-step random walks to capture high-order user similarity relationships. Treating one domain as the target, we propose a disentangled intent-wise contrastive learning approach, guided by user similarity, to refine the bridging of user intents across domains. Extensive experiments on four benchmark CDR datasets demonstrate that DisCo consistently outperforms existing state-of-the-art baselines, thereby validating the effectiveness of both DisCo and its components.

Related papers

• Coherence-guided Preference Disentanglement for Cross-domain Recommendations [71.4198675855171]
  We propose a Coherence-guided Preference Disentanglement (CoPD) method aimed at improving cross-domain recommendation by i) explicitly extracting shared item attributes to guide the learning of shared user preferences and ii) disentangling these preferences to identify specific user interests transferred between domains. Experiments conducted on real-world datasets demonstrate the superior performance of our proposed CoPD over existing competitive baselines, highlighting its effectiveness in enhancing cross-domain recommendation performance.
  arXiv  Detail & Related papers  (2024-10-27T20:21:14Z)
• Cross-Domain Sequential Recommendation via Neural Process [9.01082886458853]
  Cross-Domain Sequential Recommendation (CDSR) is a hot topic in sequence-based user interest modeling. We show how to unleash the potential of non-overlapped users' behaviors to empower CDSR.
  arXiv  Detail & Related papers  (2024-10-17T14:22:57Z)
• Adaptive Coordinators and Prompts on Heterogeneous Graphs for Cross-Domain Recommendations [31.05975545409408]
  We develop HAGO, a framework to integrate multi-domain graphs into a cohesive structure. We also develop a universal multi-domain graph pre-training strategy. Our solutions outperform state-of-the-art methods in multi-domain recommendation scenarios.
  arXiv  Detail & Related papers  (2024-10-15T15:50:53Z)
• Cross-domain Transfer of Valence Preferences via a Meta-optimization Approach [17.545983294377958]
  CVPM formalizes cross-domain interest transfer as a hybrid architecture of meta-learning and self-supervised learning. With deep insights into user preferences, we employ differentiated encoders to learn their distributions. In particular, we treat each user's mapping as two parts, the common transformation and the personalized bias, where the network used to generate the personalized bias is output by a meta-learner.
  arXiv  Detail & Related papers  (2024-06-24T10:02:24Z)
• Information Maximization via Variational Autoencoders for Cross-Domain Recommendation [26.099908029810756]
  We introduce a new CDSR framework named Information Maximization Variational Autoencoder (textbftextttIM-VAE) Here, we suggest using a Pseudo-Sequence Generator to enhance the user's interaction history input for downstream fine-grained CDSR models. To the best of our knowledge, this paper is the first CDSR work that considers the information disentanglement and denoising of pseudo-sequences in the open-world recommendation scenario.
  arXiv  Detail & Related papers  (2024-05-31T09:07:03Z)
• Exploiting Graph Structured Cross-Domain Representation for Multi-Domain Recommendation [71.45854187886088]
  Multi-domain recommender systems benefit from cross-domain representation learning and positive knowledge transfer. We use temporal intra- and inter-domain interactions as contextual information for our method called MAGRec. We perform experiments on publicly available datasets in different scenarios where MAGRec consistently outperforms state-of-the-art methods.
  arXiv  Detail & Related papers  (2023-02-12T19:51:32Z)
• Cross-domain recommendation via user interest alignment [20.387327479445773]
  Cross-domain recommendation aims to leverage knowledge from multiple domains to alleviate the data sparsity and cold-start problems in traditional recommender systems. The general practice of this approach is to train user embeddings in each domain separately and then aggregate them in a plain manner. We propose a novel cross-domain recommendation framework, namely COAST, to improve recommendation performance on dual domains.
  arXiv  Detail & Related papers  (2023-01-26T23:54:41Z)
• Diverse Preference Augmentation with Multiple Domains for Cold-start Recommendations [92.47380209981348]
  We propose a Diverse Preference Augmentation framework with multiple source domains based on meta-learning. We generate diverse ratings in a new domain of interest to handle overfitting on the case of sparse interactions. These ratings are introduced into the meta-training procedure to learn a preference meta-learner, which produces good generalization ability.
  arXiv  Detail & Related papers  (2022-04-01T10:10:50Z)
• A cross-domain recommender system using deep coupled autoencoders [77.86290991564829]
  Two novel coupled autoencoder-based deep learning methods are proposed for cross-domain recommendation. The first method aims to simultaneously learn a pair of autoencoders in order to reveal the intrinsic representations of the items in the source and target domains. The second method is derived based on a new joint regularized optimization problem, which employs two autoencoders to generate in a deep and non-linear manner the user and item-latent factors.
  arXiv  Detail & Related papers  (2021-12-08T15:14:26Z)
• Dual Metric Learning for Effective and Efficient Cross-Domain Recommendations [85.6250759280292]
  Cross domain recommender systems have been increasingly valuable for helping consumers identify useful items in different applications. Existing cross-domain models typically require large number of overlap users, which can be difficult to obtain in some applications. We propose a novel cross-domain recommendation model based on dual learning that transfers information between two related domains in an iterative manner.
  arXiv  Detail & Related papers  (2021-04-17T09:18:59Z)
• CATN: Cross-Domain Recommendation for Cold-Start Users via Aspect Transfer Network [49.35977893592626]
  We propose a cross-domain recommendation framework via aspect transfer network for cold-start users (named CATN) CATN is devised to extract multiple aspects for each user and each item from their review documents, and learn aspect correlations across domains with an attention mechanism. On real-world datasets, the proposed CATN outperforms SOTA models significantly in terms of rating prediction accuracy.
  arXiv  Detail & Related papers  (2020-05-21T10:05:19Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.