Graph Learning based Recommender Systems: A Review

Wang, Shoujin; Hu, Liang; Wang, Yan; He, Xiangnan; Sheng, Quan Z.; Orgun, Mehmet A.; Cao, Longbing; Ricci, Francesco; Yu, Philip S.

doi:10.48550/arxiv.2105.06339

Cited by 10 publications

(19 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A clustering objective function alone is incapable of achieving the optimal performance [20,26], since it is unable to capture local information such as the connectivity between two nodes u, v. Therefore, an instance-wise objective function is usually used together with the clustering objective function. In this paper, we use an objective function, shown in Equation (2), based on the popular InfoNCE [27] as the instance-wise objective function, which pulls connected u, v together and pushes away unconnected u, v in the embedding space, as illustrated in the middle part of Figure 1.…”

Section: Co-cluster Infomaxmentioning

confidence: 99%

“…The bipartite graph is a powerful representation formalism to model interactions between two types of nodes, which has been used in a variety of real-world applications. For example, in recommender systems [1,2], users, items and their interactions (e.g. buy) is a natural bipartite graph; in information retrieval [3,4], clickthrough between queries and webpages can be conveniently modeled by a bipartite graph; in drug discovery [5,6], chemical interactions (e.g.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

COIN: Co-Cluster Infomax for Bipartite Graphs

Jing¹,

Yao²,

Zhu³

et al. 2022

Preprint

View full text Add to dashboard Cite

Bipartite graphs are powerful data structures to model interactions between two types of nodes, which have been used in a variety of applications, such as recommender systems, information retrieval, and drug discovery. A fundamental challenge for bipartite graphs is how to learn informative node embeddings. Despite the success of recent self-supervised learning methods on bipartite graphs, their objectives are discriminating instance-wise positive and negative node pairs, which could contain cluster-level errors. In this paper, we introduce a novel cocluster infomax (COIN) framework, which captures the cluster-level information by maximizing the mutual information of co-clusters. Different from previous infomax methods which estimate mutual information by neural networks, COIN could easily calculate mutual information. Besides, COIN is an end-to-end coclustering method which can be trained jointly with other objective functions and optimized via back-propagation. Furthermore, we also provide theoretical analysis for COIN. We theoretically prove that COIN is able to effectively maximize the mutual information of node embeddings and COIN is upper-bounded by the prior distributions of nodes. We extensively evaluate the proposed COIN framework on various benchmark datasets and tasks to demonstrate the effectiveness of COIN.Preprint. Under review.

show abstract

Section: Co-cluster Infomaxmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

COIN: Co-Cluster Infomax for Bipartite Graphs

Jing¹,

Yao²,

Zhu³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…In [25] a GNN characterizes and predicts the local correlation structure of images in the feature space, using which, neighboring images collaborate and refine their feature embeddings based on local linear combination. The representation capability of graphs has led to various graphbased image recommendation systems also [26]. Table I gives a performance comparison of some of the results obtained in CBIR.…”

Section: Related Workmentioning

confidence: 99%

Analysis of Content Based Image Retrieval using Deep Feature Extraction and Similarity Matching

Mathews¹,

Sejal²,

Venugopal³

2022

IJACSA

View full text Add to dashboard Cite

Image retrieval using a textual query becomes a major challenge mainly due to human perception subjectivity and the impreciseness of image annotations. These drawbacks can be overcome by focusing on the content of images rather than on the textual descriptions of images. Traditional feature extraction techniques demand for expert knowledge to select the limited feature types and are also sensitive to changing imaging conditions. Deep feature extraction using Convolutional Neural Network (CNN) are a solution to these drawbacks as they can learn the feature representations automatically. This work carries out a detailed performance comparison of various pretrained models of CNN in feature extraction. Features are extracted from men footwear and women clothing datasets using the VGG16, VGG19, InceptionV3, Xception and ResNet50 models. Further, these extracted features are used for classification using SVM, Random Forest and K-Nearest Neighbors classifiers. Results of feature extraction and image retrieval show that VGG19, Inception and Xception features perform well with feature extraction, achieving a good image classification accuracy of 97.5%. These results are further justified by performing a comparison of image retrieval efficiency, with the extracted features and similarity metrics. This work also compares the accuracy obtained by features extracted by the selected pre-trained CNN models with the results obtained using conventional classification techniques on CIFAR 10 dataset. The features extracted using CNN can be used in image-based systems like recommender systems, where images have to be analyzed to generate item profiles.

show abstract

“…Recently, graph neural networks (GNN) have been adopted to build recommender systems [5]. In comparison to the classic matrix factorization methods, they consider highorder proximity between users and items [6].…”

Section: Introductionmentioning

confidence: 99%

Exploration in Sequential Recommender Systems via Graph Representations

Kiselev

Makarov

2022

IEEE Access

View full text Add to dashboard Cite

Temporal graph networks are powerful tools for solving the cold-start problem in sequential recommender systems. However, graph models are still susceptible to feedback loops and data distribution shifts. The paper proposes a simple yet efficient graph-based exploration method for the mitigation of the aforementioned issues. It adopts the counter-based state exploration from reinforcement learning to the bipartite graph domain. We suggest a method that biases model predictions using Rooted PageRank towards locally unexplored items. The method shows competitive quality on the popular recommender systems benchmarks. We, also, provide an extensive qualitative analysis of experiment results and recommendations for the use of our method in production applications.

show abstract

Graph Learning based Recommender Systems: A Review

Cited by 10 publications

References 55 publications

COIN: Co-Cluster Infomax for Bipartite Graphs

COIN: Co-Cluster Infomax for Bipartite Graphs

Analysis of Content Based Image Retrieval using Deep Feature Extraction and Similarity Matching

Exploration in Sequential Recommender Systems via Graph Representations

Contact Info

Product

Resources

About