What is the main reason why the authors used the network embedding techniques?

Using the network embedding techniques can help compress the raw input feature vector to a low-dimensional and dense vector, then the model can be easy to train.

What is the aggregation function for arithmetic?

One natural aggregation function for Aддreneiдbhors is also the mean operator which take the element-wise mean of the vectors in{

How can GraphRec capture the heterogeneity in aggregation operations?

To sum up, GraphRec can capture the heterogeneity in aggregation operations of the proposed framework via attention mechanisms, which can boost the recommendation performance.

What is the description of the deep neural network?

In recent years, deep neural network models had a great impact on learning effective feature representations in various fields, such as speech recognition [12], Computer Vision (CV) [14] and Natural Language Processing (NLP) [4].

How many epochs did the authors stop training?

The early stopping strategy was performed, where the authors stopped training if the RMSE on validation set increased for 5 successive epochs.

What are the two social recommendation models?

PMF and NeuMF are pure collaborative filtering model without social network information for rating prediction, while the others are social recommendation.

What is the main purpose of the model?

This model employs a deep neural network to learn representations of each user from social relations, and to integrate into probabilistic matrix factorization for rating prediction.

What are the main criteria used to evaluate the performance of the GraphRec?

To evaluate the performance, the authors compared their GraphRec with three groups of methods including traditional recommender systems, traditional social recommender systems, and deep neural network based recommender systems.

What is the simplest way to model the user latent factor?

the user latent factor hi is defined as,c1 = [ hIi ⊕ h S i ] (12) c2 = σ (W2 · c1 + b2) (13) ...hi = σ (Wl · cl−1 + bl ) (14)where l is the index of a hidden layer.

Open AccessProceedings Article10.1145/3308558.3313488

Graph Neural Networks for Social Recommendation

Wenqi Fan, +6 more

- 13 May 2019

- pp 417-426

Chat with Paper

AI Agents for this Paper

Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps

Most frequently asked questions

1. What are the contributions mentioned in the paper "Graph neural networks for social recommendation" ?

To address the three aforementioned challenges simultaneously, in this paper, the authors present a novel graph neural network framework ( GraphRec ) for social recommendations.. In particular, the authors provide a principled approach to jointly capture interactions and opinions in the user-item graph and propose the framework GraphRec, which coherently models two graphs and heterogeneous strengths.

2. What future works have the authors mentioned in the paper "Graph neural networks for social recommendation" ?

Therefore, exploring graph neural networks for recommendation with attributes would be an interesting future direction.

3. How did SoDimRec first use social relations?

SoDimRec [30] first adopted a community detection algorithm to partition users into several clusters, and then exploited the heterogeneity of social relations and weak dependency connections for recommendation.

4. What are the popular metrics used to evaluate the quality of the recommendation algorithms?

In order to evaluate the quality of the recommendation algorithms, two popular metrics are adopted to evaluate the predictive accuracy, namely Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) [34].

Figure 5: Effect of embedding size on Ciao and Epinions datasets.

Figure 1: Graph Data in Social Recommendation. It contains two graphs including the user-item graph (left part) and the user-user social graph (right part). Note that the number on the edges of the user-item graph denotes the opinions (or rating score) of users on the items via the interactions.

Table 3: Performance comparison of different recommender systems

Figure 2: The overall architecture of the proposed model. It contains three major components: user modeling, item modeling, and rating prediction.

Citations

Journal Article•10.1145/3616542

Faithful and Consistent Graph Neural Network Explanations with Rationale Alignment

Tianxiang Zhao, +3 more

- 09 Oct 2023

- ACM Transactions on Intelligent Systems ...

TL;DR: Faithful and consistent GNN explanations with rationale alignment. The existing GNN explanation methods based on minimal subgraph search suffer from inductive bias and provide spurious explanations. This paper proposes a new explanation framework with an auxiliary alignment loss that optimizes a more faithful explanation objective intrinsically.

...read moreread less

Journal Article•10.1109/tsc.2022.3180163

H3Rec: Higher-Order Heterogeneous and Homogeneous Interaction Modeling for Group Recommendations of Web Services

01 Mar 2023

- IEEE Transactions on Services Computing

TL;DR: Wang et al. as discussed by the authors proposed a new model, called H3Rec, which learns the representations of entities by developing two graph embedding layers based on an interaction graph of all entities.

...read moreread less

Journal Article•10.1109/tnnls.2023.3272475

Diversifying Collaborative Filtering via Graph Spreading Network and Selective Sampling

01 Jan 2023

- IEEE transactions on neural networks and...

TL;DR: In this paper , a graph spreading network (GSN) is proposed to address the accuracy-diversity dilemma of CF. But, it suffers from the difficulty of adapting to different scenarios' demands concerning the accuracydiversity ratio of their recommendation lists.

...read moreread less

Journal Article•10.1007/s10586-024-04560-x

HyperSegRec: enhanced hypergraph-based recommendation system with user segmentation and item similarity learning

Nidhi Malik, +5 more

- 03 Jun 2024

- Cluster Computing

Journal Article•10.48550/arxiv.2403.04747

GNN-VPA: A Variance-Preserving Aggregation Strategy for Graph Neural Networks

Lisa Schneckenreiter, +5 more

- 07 Mar 2024

- arXiv.org

TL;DR: GNN-VPA introduces a variance-preserving aggregation function that enhances expressivity and learning dynamics in GNNs.

...read moreread less

...

Expand

References

•Journal Article

Dropout: a simple way to prevent neural networks from overfitting

Nitish Srivastava, +4 more

- 01 Jan 2014

- Journal of Machine Learning Research

TL;DR: It is shown that dropout improves the performance of neural networks on supervised learning tasks in vision, speech recognition, document classification and computational biology, obtaining state-of-the-art results on many benchmark data sets.

...read moreread less

43.7K

•Posted Content

Semi-Supervised Classification with Graph Convolutional Networks

Thomas Kipf, +1 more

- 09 Sep 2016

- arXiv: Learning

TL;DR: A scalable approach for semi-supervised learning on graph-structured data that is based on an efficient variant of convolutional neural networks which operate directly on graphs which outperforms related methods by a significant margin.

...read moreread less

22.7K

Journal Article•10.1146/ANNUREV.SOC.27.1.415

Birds of a Feather: Homophily in Social Networks

Miller McPherson, +2 more

- 01 Aug 2001

- Review of Sociology

TL;DR: The homophily principle as mentioned in this paper states that similarity breeds connection, and that people's personal networks are homogeneous with regard to many sociodemographic, behavioral, and intrapersonal characteristics.

...read moreread less

19.1K

•Posted Content

Inductive Representation Learning on Large Graphs

William L. Hamilton, +2 more

- 07 Jun 2017

- arXiv: Social and Information Networks

TL;DR: GraphSAGE is presented, a general, inductive framework that leverages node feature information (e.g., text attributes) to efficiently generate node embeddings for previously unseen data and outperforms strong baselines on three inductive node-classification benchmarks.

...read moreread less

11.9K

Journal Article•10.1109/MSP.2012.2205597

Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research Groups

Geoffrey E. Hinton, +10 more

- 18 Oct 2012

- IEEE Signal Processing Magazine

TL;DR: This article provides an overview of progress and represents the shared views of four research groups that have had recent successes in using DNNs for acoustic modeling in speech recognition.

...read moreread less

11.4K

...

Expand

Graph Neural Networks for Social Recommendation

Chat with Paper

AI Agents for this Paper

Most frequently asked questions

1. What are the contributions mentioned in the paper "Graph neural networks for social recommendation" ?

2. What future works have the authors mentioned in the paper "Graph neural networks for social recommendation" ?

3. How did SoDimRec first use social relations?

4. What are the popular metrics used to evaluate the quality of the recommendation algorithms?

5. What is the main reason why the authors used the network embedding techniques?

6. What is the aggregation function for arithmetic?

7. How can GraphRec capture the heterogeneity in aggregation operations?

8. What is the description of the deep neural network?

9. How many epochs did the authors stop training?

10. What are the two social recommendation models?

11. What is the main purpose of the model?

12. What are the main criteria used to evaluate the performance of the GraphRec?

13. What is the simplest way to model the user latent factor?

Figures

Citations

Faithful and Consistent Graph Neural Network Explanations with Rationale Alignment

H3Rec: Higher-Order Heterogeneous and Homogeneous Interaction Modeling for Group Recommendations of Web Services

Diversifying Collaborative Filtering via Graph Spreading Network and Selective Sampling

HyperSegRec: enhanced hypergraph-based recommendation system with user segmentation and item similarity learning

GNN-VPA: A Variance-Preserving Aggregation Strategy for Graph Neural Networks

References

Dropout: a simple way to prevent neural networks from overfitting

Semi-Supervised Classification with Graph Convolutional Networks

Birds of a Feather: Homophily in Social Networks

Inductive Representation Learning on Large Graphs

Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research Groups

Related Papers (5)

Inductive Representation Learning on Large Graphs

Neural Collaborative Filtering

Graph Attention Networks

Semi-Supervised Classification with Graph Convolutional Networks

Matrix Factorization Techniques for Recommender Systems