Cross-domain Recommendation Research Articles

Sequential and Graphical Cross-Domain Recommendations with a Multi-View Hierarchical Transfer Gate

Cross-domain recommender systems could potentially improve the recommendation performance by means of transferring abundant knowledge from the auxiliary domain to the target domain. They could help address some key challenges in recommender systems, such as data sparsity and cold start. However, most existing cross-domain recommendation approaches represent the user preferences based on a single kind of user’s feature or behavior and fail to explore the hidden interaction effects of different kinds of features or behaviors. In this article, we propose the S equential and G raphical Cross -Domain Recommendations with a Multi-View Hierarchical Transfer Gate (SGCross) to transfer user representations from multiple perspectives. The SGCross model constructs a user profile by learning the personal preference from a personal view, the dynamic preference from a temporal view, as well as the collaborative preference from a collaborative view. Specifically, a Multi-view Hierarchical Gate (MHG) is designed to transfer the informative representations of user knowledge on different views from the auxiliary domain separately, aiming to enhance the user representations. Furthermore, a two-stage attentive fusion module is designed to integrate transferred information at two levels: the domain level and the view level. Extensive experiments on the Amazon dataset and the Douban dataset have demonstrated that SGCross effectively improves the accuracy of cross-domain recommendations and outperforms the state-of-the-art baseline models.

CDR-ULPT:Cross-domain recommendation for e-commerce using user-level preferences transfer network

Recommender systems have been widely used on e-commerce websites because they can assist users in finding products that fit their needs among the masses of products. However, recommender systems face the problem of data sparsity. Cross-domain recommender systems can alleviate this question and have been a hot research area. Adversarial domain adaptation has been a hot topic in cross-domain recommender systems due to its ability to mitigate the differences in feature distribution among different domains. Recent research has focused on learning users' domain sharing and private preferences. However, adversarial domain adaptation models ignore the individualized personal distinctions of users' domain-sharing preferences in the alignment process. In addition, these models missed indirect interaction. Graph convolutional model can mine the indirect interaction but ignore the different influences from neighboring users. Therefore, this paper proposes a Cross-domain Recommendation for e-commerce using the User-level Preferences Transfer Network (CDR-ULPT). The main work is as follows: (1) Adding an attention mechanism on adversarial domain adaptation to personalize the alignment of users' domain-sharing preferences. (2) Capturing users' indirect interactions using multiple graph convolution layers, adding node attention on the graph convolution layer to dynamically aggregate neighbor features. Experiments on five public datasets demonstrate that the method proposed in this work outperforms other cross-domain recommendation methods.

A Cross-Domain Multimodal Supervised Latent Topic Model for Item Tagging and Cold-Start Recommendation

Cross-domain data analysis is playing an increasingly important role in media convergence and can be adopted for many applications. Most existing methods consider the domain discrimination as the multimodal representation difference or the imbalanced item classification distribution, which ignore the different tag semantics between domains. To this end, we propose an explainable cross-domain multimodal supervised latent topic model (CDMSLT) and evaluate our model on two applications. First, we learn a common topic space which is capable of explaining both domain specification and commonality. Second, we apply our model to multilabel classification task and put forward a cross-domain item tagging method (CDMSLT-tag). Third, combining user behaviors and CDMSLT model, we propose a cross-domain recommendation algorithm (CDMSLT-rec) which could estimate the user preference on new unseen domains. This paper proves the effectiveness of CDMSLT model by comparing these two applications with existing algorithms in a cross-domain scenario on Douban dataset.

PPGenCDR: A Stable and Robust Framework for Privacy-Preserving Cross-Domain Recommendation

Privacy-preserving cross-domain recommendation (PPCDR) refers to preserving the privacy of users when transferring the knowledge from source domain to target domain for better performance, which is vital for the long-term development of recommender systems. Existing work on cross-domain recommendation (CDR) reaches advanced and satisfying recommendation performance, but mostly neglects preserving privacy. To fill this gap, we propose a privacy-preserving generative cross-domain recommendation (PPGenCDR) framework for PPCDR. PPGenCDR includes two main modules, i.e., stable privacy-preserving generator module, and robust cross-domain recommendation module. Specifically, the former isolates data from different domains with a generative adversarial network (GAN) based model, which stably estimates the distribution of private data in the source domain with ́Renyi differential privacy (RDP) technique. Then the latter aims to robustly leverage the perturbed but effective knowledge from the source domain with the raw data in target domain to improve recommendation performance. Three key modules, i.e., (1) selective privacy preserver, (2) GAN stabilizer, and (3) robustness conductor, guarantee the cost-effective trade-off between utility and privacy, the stability of GAN when using RDP, and the robustness of leveraging transferable knowledge accordingly. The extensive empirical studies on Douban and Amazon datasets demonstrate that PPGenCDR significantly outperforms the state-of-the-art recommendation models while preserving privacy.

Win-Win: A Privacy-Preserving Federated Framework for Dual-Target Cross-Domain Recommendation

Cross-domain recommendation (CDR) aims to alleviate the data sparsity by transferring knowledge from an informative source domain to the target domain, which inevitably proposes stern challenges to data privacy and transferability during the transfer process. A small amount of recent CDR works have investigated privacy protection, while they still suffer from satisfying practical requirements (e.g., limited privacy-preserving ability) and preventing the potential risk of negative transfer. To address the above challenging problems, we propose a novel and unified privacy-preserving federated framework for dual-target CDR, namely P2FCDR. We design P2FCDR as peer-to-peer federated network architecture to ensure the local data storage and privacy protection of business partners. Specifically, for the special knowledge transfer process in CDR under federated settings, we initialize an optimizable orthogonal mapping matrix to learn the embedding transformation across domains and adopt the local differential privacy technique on the transformed embedding before exchanging across domains, which provides more reliable privacy protection. Furthermore, we exploit the similarity between in-domain and cross-domain embedding, and develop a gated selecting vector to refine the information fusion for more accurate dual transfer. Extensive experiments on three real-world datasets demonstrate that P2FCDR significantly outperforms the state-of-the-art methods and effectively protects data privacy.

Self-Supervised Interest Transfer Network via Prototypical Contrastive Learning for Recommendation

Cross-domain recommendation has attracted increasing attention from industry and academia recently. However, most existing methods do not exploit the interest invariance between domains, which would yield sub-optimal solutions. In this paper, we propose a cross-domain recommendation method: Self-supervised Interest Transfer Network (SITN), which can effectively transfer invariant knowledge between domains via prototypical contrastive learning. Specifically, we perform two levels of cross-domain contrastive learning: 1) instance-to-instance contrastive learning, 2) instance-to-cluster contrastive learning. Not only that, we also take into account users' multi-granularity and multi-view interests. With this paradigm, SITN can explicitly learn the invariant knowledge of interest clusters between domains and accurately capture users' intents and preferences. We conducted extensive experiments on a public dataset and a large-scale industrial dataset collected from one of the world's leading e-commerce corporations. The experimental results indicate that SITN achieves significant improvements over state-of-the-art recommendation methods. Additionally, SITN has been deployed on a micro-video recommendation platform, and the online A/B testing results further demonstrate its practical value. Supplement is available at: https://github.com/fanqieCoffee/SITN-Supplement.

REMIT: Reinforced Multi-Interest Transfer for Cross-Domain Recommendation

Cold-start problem is one of the most challenging problems for recommender systems. One promising solution to this problem is cross-domain recommendation (CDR) which leverages rich information from an auxiliary source domain to improve the performance of recommender system in the target domain. In particular, the family of embedding and mapping methods for CDR is very effective, which explicitly learn a mapping function from source embeddings to target embeddings to transfer user’s preferences. Recent works usually transfer an overall source embedding by modeling a common or personalized preference bridge for all users. However, a unified user embedding cannot reflect the user’s multiple interests in auxiliary source domain. In this paper, we propose a novel framework called reinforced multi-interest transfer for CDR (REMIT). Specifically, we first construct a heterogeneous information network and employ different meta-path based aggregations to get user’s multiple interests in source domain, then transform different interest embeddings with different meta-generated personalized bridge functions for each user. To better coordinate the transformed user interest embeddings and the item embedding in target domain, we systematically develop a reinforced method to dynamically assign weights to transformed interests for different training instances and optimize the performance of target model. In addition, the REMIT is a general framework that can be applied upon various base models in target domain. Our extensive experimental results on large real-world datasets demonstrate the superior performance and compatibility of REMIT.

Non-IID Transfer Learning on Graphs

Transfer learning refers to the transfer of knowledge or information from a relevant source domain to a target domain. However, most existing transfer learning theories and algorithms focus on IID tasks, where the source/target samples are assumed to be independent and identically distributed. Very little effort is devoted to theoretically studying the knowledge transferability on non-IID tasks, e.g., cross-network mining. To bridge the gap, in this paper, we propose rigorous generalization bounds and algorithms for cross-network transfer learning from a source graph to a target graph. The crucial idea is to characterize the cross-network knowledge transferability from the perspective of the Weisfeiler-Lehman graph isomorphism test. To this end, we propose a novel Graph Subtree Discrepancy to measure the graph distribution shift between source and target graphs. Then the generalization error bounds on cross-network transfer learning, including both cross-network node classification and link prediction tasks, can be derived in terms of the source knowledge and the Graph Subtree Discrepancy across domains. This thereby motivates us to propose a generic graph adaptive network (GRADE) to minimize the distribution shift between source and target graphs for cross-network transfer learning. Experimental results verify the effectiveness and efficiency of our GRADE framework on both cross-network node classification and cross-domain recommendation tasks.

Heterogeneous and clustering-enhanced personalized preference transfer for cross-domain recommendation

As a promising solution to alleviate the critical cold-start problem in recommendation systems, Cross-Domain Recommendation (CDR) aims to transfer users’ preferences from the source domain to the target domain. However, the negligence of semantic differences of heterogeneous relations when modeling user preferences and the non-consideration of a user’s common characteristics inferred from similar users when transferring the user’s preferences make recent CDR works cannot well handle the two core issues of CDR—‘what to transfer’ and ‘how to transfer’. To this end, we propose a novel heterogeneous and clustering-enhanced user preference transfer model for CDR (named HCCDR). To well address the first issue, the heterogeneous latent factor modeling component is firstly built to compute high-quality representations of users and items to be transferred based on the heterogeneous relations among users and items. Note that the heterogeneous relations with different semantics are processed with different models. Then, the clustering-enhanced preference transfer component well addresses the second issue by learning an effective personalized preference transfer function between two domains for a user, where the individual and common characteristics of the user are concurrently considered. Finally, the personalized recommendation component achieves the personalized recommendation in the target domain based on the transferred user embeddings calculated via the learned personalized preference transfer functions. Experimental results performed on large public datasets demonstrate that the proposed HCCDR markedly outperforms all baselines. In particular, HCCDR gains a maximum performance improvement of 12.69% (or 8.99%) for RMSE (or MAE) compared with the best baseline.

GroupAligner : A Deep Reinforcement Learning with Domain Adaptation for Social Group Alignment

Social network alignment, which aims to uncover the correspondence across different social networks, shows fundamental importance in a wide spectrum of applications such as cross-domain recommendation and information propagation. In the literature, the vast majority of the existing studies focus on the social network alignment at user level. In practice, the user-level alignment usually relies on abundant personal information and high-quality supervision, which is expensive and even impossible in the real-world scenario. Alternatively, we propose to study the problem of social group alignment across different social networks, focusing on the interests of social groups rather than personal information. However, social group alignment is non-trivial and faces significant challenges in both (i) feature inconsistency across different social networks and (ii) group discovery within a social network. To bridge this gap, we present a novel GroupAligner , a deep reinforcement learning with domain adaptation for social group alignment. In GroupAligner , to address the first issue, we propose the cycle domain adaptation approach with the Wasserstein distance to transfer the knowledge from the source social network, aligning the feature space of social networks in the distribution level. To address the second issue, we model the group discovery as a sequential decision process with reinforcement learning in which the policy is parameterized by a proposed p roximity-enhanced G raph N eural N etwork (pGNN) and a GNN-based discriminator to score the reward. Finally, we utilize pre-training and teacher forcing to stabilize the learning process of GroupAligner . Extensive experiments on several real-world datasets are conducted to evaluate GroupAligner , and experimental results show that GroupAligner outperforms the alternative methods for social group alignment.

Dual attentive graph convolutional networks for cross-domain recommendation

Cross-domain recommendation aims to alleviate the target domain’s data sparsity problem by leveraging source domain knowledge. Existing GCN-based approaches perform graph convolution operations in each domain separately. However, the direct effect of item feature and topological structure information in the source domain are neglected for user preference modeling in the target domain. In this paper, we propose a novel Dual Attentive Graph Convolutional Network for Cross-Domain Recommendation (DAG4CDR). Specifically, we integrate the source and target domain’s interaction data to construct a unified user-item bipartite graph and then perform GCN propagation on the graph to learn user and item embeddings. Over the unified graph, the interaction data from both domains can be leveraged to learn user and item embeddings via information propagation. In the embedding aggregation phase, the messages passed from different items of two domains to users are weighted by a designed dual attention mechanism, which considers the contributions of different items from both node- and domain-level. We conducted extensive experiments to validate the effectiveness of our method on several publicly available datasets, and the results demonstrate the superiority of our model on preference modeling for both common and non-common users.

DaGzang: a synthetic data generator for cross-domain recommendation services.

Research on cross-domain recommendation systems (CDRS) has shown efficiency by leveraging the overlapping associations between domains in order to generate more encompassing user models and better recommendations. Nonetheless, if there is no dataset belonging to a specific domain, it is a challenge to generate recommendations in CDRS. In addition, finding these overlapping associations in the real world is generally tricky, and it makes its application to actual services hard. Considering these issues, this study aims to present a synthetic data generation platform (called DaGzang) for cross-domain recommendation systems. The DaGzang platform works according to the complete loop, and it consists of the following three steps: (i) detecting the overlap association (data distribution pattern) between the real-world datasets, (ii) generating synthetic datasets based on these overlap associations, and (iii) evaluating the quality of the generated synthetic datasets. The real-world datasets in our experiments were collected from Amazon's e-commercial website. To validate the usefulness of the synthetic datasets generated from DaGzang, we embed these datasets into our cross-domain recommender system, called DakGalBi. We then evaluate the recommendations generated from DakGalBi with collaborative filtering (CF) algorithms, user-based CF, and item-based CF. Mean absolute error (MAE) and root mean square error (RMSE) metrics are measured to evaluate the performance of collaborative filtering (CF) CDRS. In particular, the highest performance of the three recommendation methods is user-based CF when using 10 synthetic datasets generated from DaGzang (0.437 at MAE and 0.465 at RMSE).

Cross-domain recommendation model of deep feature extraction and attention mechanism

Cross-domain recommendation model of deep feature extraction and attention mechanism

A Deep Dual Adversarial Network for Cross-Domain Recommendation

A Deep Dual Adversarial Network for Cross-Domain Recommendation

MuKGB-CRS: Guarantee privacy and authenticity of cross-domain recommendation via multi-feature knowledge graph integrated blockchain

In distributed cross-domain recommendation systems, current privacy protection techniques are ineffective at verifying data correctness and existing authenticity techniques have limitations in privacy protection. This poses a conflict between security and credibility. This paper designs a trusted cross-domain recommendation model based on multi-feature knowledge graph (MuKG) and blockchain to boost recommendation accuracy, data credibility and security. To the best of our knowledge, MuKG is proposed, for the first time, to realize a unified representation of multi-source heterogeneous data in a secure manner. Under this model, federated learning integrated with blockchain is used to implement a co-trust mechanism for distributed learning while guaranteeing the credibility of data through blockchain traceability. Our model can guarantee both data security and authenticity with no need of generalization for privacy protection and coordination of centralized servers for distributed controls. Experiments are performed on the two classic datasets, MovieLens and Amazon, with different sparsity. The results have shown that MuKG improves the recommendation accuracy by 1.5 % and diversity by 18 % while ensuring data security and credibility in sparse data. We also verify that different data characteristics have different influence on the recommendation results.

MFPCDR: A Meta-Learning-Based Model for Federated Personalized Cross-Domain Recommendation

Cross-domain recommendation systems frequently require the use of rich source domain information to improve recommendations in the target domain, thereby resolving the data sparsity and cold-start problems, whereas the majority of existing approaches frequently require the centralized storage of user data, which poses a substantial risk of privacy breaches. Compared to traditional recommendation systems with centralized data, federated recommendation systems with multiple clients trained collaboratively have significant privacy benefits in terms of user data. While users’ interests are often personalized, meta-learning can be used to learn users’ personalized preferences, and personalized preferences can help models make recommendations in cold-start scenarios. We use meta-learning to learn the personalized preferences of cold-start users. Therefore, we offer a unique meta-learning-based federated personalized cross-domain recommendation model that discovers the personalized preferences for cold-start users via a server-side meta-recommendation module. To avoid compromising user privacy, an attention mechanism is used on each client to find transferable features that contribute to knowledge transfer while obtaining embeddings of users and items; each client then uploads the weights to the server. The server accumulates weights and delivers them to clients for update. Compared to traditional recommendation system models, our model can effectively protect user privacy while solving the user cold-start problem, as we use an attention mechanism in the local embedding module to mine the source domain for transferable features that contribute to knowledge transfer. Extensive trials on real-world datasets have demonstrated that our technique effectively guarantees speed while protecting user privacy.

Multi-level Attention-based Domain Disentanglement for BCDR

Cross-domain recommendation aims to exploit heterogeneous information from a data-sufficient domain (source domain) to transfer knowledge to a data-scarce domain (target domain). A majority of existing methods focus on unidirectional transfer that leverages the domain-shared information to facilitate the recommendation of the target domain. Nevertheless, it is more beneficial to improve the recommendation performance of both domains simultaneously via a dual transfer learning schema, which is known as bidirectional cross-domain recommendation (BCDR). Existing BCDR methods have their limitations, since they only perform bidirectional transfer learning based on domain-shared representations while neglecting rich information that is private to each domain. In this article, we argue that users may have domain-biased preferences due to the characteristics of that domain. Namely, the domain-specific preference information also plays a critical role in the recommendation. To effectively leverage the domain-specific information, we propose a M ulti-level A ttention-based D omain D isentanglement framework dubbed MADD for BCDR, which explicitly leverages the attention mechanism to construct personalized preference with both domain-invariant and domain-specific features obtained by disentangling raw user embeddings. Specifically, the domain-invariant feature is exploited by domain-adversarial learning while the domain-specific feature is learned by imposing an orthogonal loss. We then conduct a reconstruction process on disentangled features to ensure semantic-sufficiency. After that, we devise a multi-level attention mechanism for these disentangled features, which determines their contributions to the final personalized user preference embedding by dynamically learning the attention scores of individual features. We train the model in a multi-task learning fashion to benefit both domains. Extensive experiments on real-world datasets demonstrate that our model significantly outperforms state-of-the-art cross-domain recommendation approaches.

Disentangled Representations Learning for Multi-target Cross-domain Recommendation

Data sparsity has been a long-standing issue for accurate and trustworthy recommendation systems (RS). To alleviate the problem, many researchers pay much attention to cross-domain recommendation (CDR), which aims at transferring rich knowledge from related source domains to enhance the recommendation performance of sparse target domain. To reach the knowledge transferring purpose, recent CDR works always focus on designing different pairwise directed or undirected information transferring strategies between source and target domains. However, such pairwise transferring idea is difficult to adapt to multi-target CDR scenarios directly, e.g., transferring knowledge between multiple domains and improving their performance simultaneously, as such strategies may lead the following issues: (1) When the number of domains increases, the number of transferring modules will grow exponentially, which causes heavy computation complexity. (2) A single pairwise transferring module could only capture the relevant information of two domains, but ignores the correlated information of other domains, which may limit the transferring effectiveness. (3) When a sparse domain serves as the source domain during the pairwise transferring, it would easily leads the negative transfer problem, and the untrustworthy information may hurt the target domain recommendation performance. In this article, we consider the key challenge of the multi-target CDR task: How to identify the most valuable trustworthy information over multiple domains and transfer such information efficiently to avoid the negative transfer problem? To fulfill the above challenge, we propose a novel end-to-end model termed as DR-MTCDR , standing for D isentangled R epresentations learning for M ulti- T arget CDR . DR-MTCDR aims at transferring the trustworthy domain-shared information across domains, which has the two major advantages in both efficiency and effectiveness: (1) For efficiency, DR-MTCDR utilizes a unified module on all domains to capture disentangled domain-shared information and domain-specific information, which could support all domain recommendation and be insensitive to the number of domains. (2) For effectiveness, based on the disentangled domain-shared and domain-specific information, DR-MTCDR has the capability to lead positive effect and make trustworthy recommendation for each domain. Empirical evaluations on datasets from both public datasets and real-world large-scale financial datasets have shown that the proposed framework outperforms other state-of-the-art baselines.

Cross-domain recommendation with Multi-Auxiliary Domains via Consistent and Selective Cluster-Level Knowledge Transfer

Recommender systems are software tools and techniques designed to provide users with useful items suggestions. For most existing recommender systems, data sparsity is a challenging and common problem. The cross-domain recommender systems can mitigate the sparsity problem of data in a target domain by introducing auxiliary domain with relatively dense data. However, there are three challenging issues in the existing cross-domain recommender systems: (1) Most cross-domain recommender systems are designed for a single auxiliary domain, which may lead to insufficient knowledge transfer. However, in practice, there may be multiple auxiliary domains related to the target domain that are available; (2) Due to the distribution divergence between the auxiliary and target domains, inconsistent knowledge may be transferred to the target domain; (3) Because of the diversity among domains, they do not necessarily share exactly the same rating pattern. Moreover, not all cluster-level knowledge of auxiliary domains is beneficial to the target domain, and some cluster-level knowledge may be harmful to the target domain. To deal with the above three issues, we propose a Cross-Domain Recommendation with Multi-Auxiliary Domains via Consistent and Selective Cluster-Level Knowledge Transfer, called MD-CSKT. First, rating matrices from multiple auxiliary domains are decomposed jointly with a single target rating matrix to transfer richer knowledge from multiple auxiliary domains. Secondly, the Maximum Mean Discrepancy (MMD) regularization constraint is incorporated into the joint decomposition to reduce the distribution discrepancy between each auxiliary domain and the target domain, so as to achieve consistent cluster-level knowledge transfer. Thirdly, selective cluster-level knowledge transfer is realized by adaptively selecting of cluster-level rating knowledge from multiple auxiliary domains in the joint matrix factorization. The experimental results on six real-world datasets in three categories demonstrate that the presented MD-CSKT method substantially outperforms nine comparison methods and achieves the improvement of recommendation accuracy in the target domain. Compared with the state-of-the-art comparison methods, the overall performance of the MD-CSKT method is improved by more than 25%.

Preference-aware Graph Attention Networks for Cross-Domain Recommendations with Collaborative Knowledge Graph

Knowledge graphs (KGs) can provide users with semantic information and relations among numerous entities and nodes, which can greatly facilitate the performance of recommender systems. However, existing KG-based approaches still suffer from severe data sparsity and may not be effective in capturing the preference features of similar entities across domains. Therefore, in this article, we propose a P reference-aware G raph A ttention network model with C ollaborative K nowledge G raph ( PGACKG ) for cross-domain recommendations. Preference-aware entity embeddings with some collaborative signals are first obtained by exploiting the graph-embedding model, which can transform entities and items in the collaborative knowledge graph into semantic preference spaces. To better learn user preference features, we devise a preference-aware graph attention network framework that aggregates the preference features of similar entities within domains and across domains. In this framework, multi-hop reasoning is employed to assist in the generation of preference features within domains, and the node random walk based on frequency visits is proposed to gather similar preferences across domains for target entities. Then, the final preference features of entities are fused, while a novel C ross-domain B ayesian P ersonalized R anking ( CBPR ) is proposed to improve cross-domain recommendation accuracy. Extensive empirical experiments on four real-world datasets demonstrate that our proposed approach consistently outperforms state-of-the-art baselines. Furthermore, our PGACKG achieves strong performance in different ablation scenarios, and the interaction sparsity experiments also demonstrate that our proposed approach can significantly alleviate the data sparsity issue.