Hypergraph Contrastive Collaborative Filtering

Xia, Lianghao; Huang, Chao; Xu, Yong; Zhao, Junhua; Yin, Dawei; Huang, Jimmy Xiangji

doi:10.1145/3477495.3532058

Cited by 161 publications

(57 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Dong-Kyu Chae et al [22] proposed AR-CF, which stands for Augmented Reality CF, a novel framework for addressing the cold-start problems by generating virtual, but plausible neighbours for cold-start users or items and augmenting them to the rating matrix as additional information for CF models. Lianghao Xia et al [23] propose a new self-…”

Section: Introductionmentioning

confidence: 99%

[Retracted] PCA and Binary K‐Means Clustering Based Collaborative Filtering Recommendation

Peng

Wang

et al. 2023

Journal of Sensors

View full text Add to dashboard Cite

Aiming at the problem of similarity calculation error caused by the extremely sparse data in collaborative filtering recommendation algorithm, a collaborative filtering recommendation algorithm based on slope one matrix prefilling model, principal component dimension reduction, and binary K -means clustering is proposed in this paper. Firstly, the algorithm uses the slope one model based on item similarity to prefill the original scoring matrix. Secondly, principal component analysis is used to reduce the dimension of the filled matrix, retain the most representative dimension of user characteristics, and remove the dimension with less information. Finally, in order to solve the time-consuming problem of similarity calculation of collaborative filtering algorithm in the case of large-scale system, binary K -means clustering is carried out in the reduced dimension vector space to reduce the search range of the nearest neighbour of the target user. The algorithm ensures the efficiency and accuracy of recommendation while the scale of users is expanded. The experimental results on movielens dataset show that the algorithm proposed in this paper is superior to the traditional collaborative filtering algorithm and the collaborative filtering recommendation algorithm based on PCA (principal component analysis) and binary K -means clustering in recall rate, accuracy rate, average error, and running time.

show abstract

Section: Introductionmentioning

confidence: 99%

[Retracted] PCA and Binary K‐Means Clustering Based Collaborative Filtering Recommendation

Peng

Wang

et al. 2023

Journal of Sensors

View full text Add to dashboard Cite

show abstract

“…Specifically, we adopt the InfoNCE loss function to maximize the mutual information between the modality-specific embeddings e 𝑚 𝑢 and the overall user embedding h 𝑢 of the same user 𝑢. With the self-discrimination strategy [46,47], embeddings from different users are treated as negative pairs. Our cross-modal contrastive loss is defined as:…”

Section: 22mentioning

confidence: 99%

“…At its core is to augment original supervision signals with the incorporated auxiliary learning task. For graph augmentation with contrastive learning, NCL [21], CML [41] and HCCF [47] propose to generate SSL signals via contrasting positive node pairs based on various augmentation operators, e.g., random walk graph sampling and semantic neighbor identification. For SSL-based sequence augmentation, CL4SRec [48] augments item sequence in three different ways, i.e., crop, mask and reorder.…”

Section: Related Workmentioning

confidence: 99%

Multi-Modal Self-Supervised Learning for Recommendation

Wei,

Huang,

Xia

et al. 2023

Preprint

View full text Add to dashboard Cite

The online emergence of multi-modal sharing platforms (e.g., Tik-Tok, Youtube) is powering personalized recommender systems to incorporate various modalities (e.g., visual, textual and acoustic) into the latent user representations. While existing works on multimodal recommendation exploit multimedia content features in enhancing item embeddings, their model representation capability is limited by heavy label reliance and weak robustness on sparse user behavior data. Inspired by the recent progress of self-supervised learning in alleviating label scarcity issue, we explore deriving selfsupervision signals with effectively learning of modality-aware user preference and cross-modal dependencies. To this end, we propose a new Multi-Modal Self-Supervised Learning (MMSSL) method which tackles two key challenges. Specifically, to characterize the inter-dependency between the user-item collaborative view and item multi-modal semantic view, we design a modality-aware interactive structure learning paradigm via adversarial perturbations for data augmentation. In addition, to capture the effects that user's modality-aware interaction pattern would interweave with each other, a cross-modal contrastive learning approach is introduced to jointly preserve the inter-modal semantic commonality and user preference diversity. Experiments on real-world datasets verify the superiority of our method in offering great potential for multimedia recommendation over various state-of-the-art baselines. The implementation is released at: https://github.com/HKUDS/MMSSL.

show abstract

“…SGL [27] explores self-supervised learning on user-item graph to alleviate the long tail effect. Through hypergraph-enhanced cross-view contrast learning architecture, HCCF [28] can jointly capture local and global cooperative relationships, enhance the recognition ability of the CF paradigm based on GNN, and comprehensively capture the complex higher-order dependency relationships between users, effectively combining hypergraph structure coding with self-supervised learning. BC-Loss [30] incorporates the perceived margin of deviation into comparison losses, where the margin is quantitatively adjusted for the degree of deviation in each user-item interaction.…”

Section: Related Workmentioning

confidence: 99%

IcaGCN: Model Intents via Coactivated Graph Convolution Network for Recommendation

Zhang

Yang

2023

IEEE Access

View full text Add to dashboard Cite

In this era of information overload, to better provide personalized content services to users, recommendation systems have greatly improved the efficiency of information distribution. Graph Convolution Network(GCN), which is one of the representative works of graph structure aggregation processing, works by node convolution with the help of the Laplacian matrix of the graph and weighted combination of neighbor node information according to the outgoing and incoming degrees of neighbor nodes to obtain the representation of the current node. However, the mainstream GCN models nowadays do not take into account data augmentation of metadata and the fact that each node plays different roles with different importance and weights, thus making the recommendation performance limited. To better solve the above problems, we propose the IcaGCN model, which can perform data augmentation and calculate node weights in modules, and is a convenient plug-and-play method. Finally, extensive experimental results on four real-world datasets have shown the effectiveness and robustness of the proposed model. Especially on the Amazon-Book dataset, our IcaGCN has improved by 6.32%, 42.29%, and 12.38% in Recall@20, MRR@20, and NDCG@20, respectively, compared to other existing state-of-the-art models. We also provide source code and data to reproduce the experimental results a .

show abstract

Hypergraph Contrastive Collaborative Filtering

Cited by 161 publications

References 25 publications

[Retracted] PCA and Binary K‐Means Clustering Based Collaborative Filtering Recommendation

[Retracted] PCA and Binary K‐Means Clustering Based Collaborative Filtering Recommendation

Multi-Modal Self-Supervised Learning for Recommendation

IcaGCN: Model Intents via Coactivated Graph Convolution Network for Recommendation

Contact Info

Product

Resources

About