A survey of graph neural networks in various learning paradigms: methods, applications, and challenges

Waikhom, Lilapati; Patgiri, Ripon

doi:10.1007/s10462-022-10321-2

Cited by 43 publications

(12 citation statements)

References 216 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Section: Data Description and Methodsmentioning

confidence: 99%

“…GNNs are commonly employed for processing graph data, which are composed of nodes and edges connecting the nodes, denoted as G = (V, E), where V and E represent the nodes and edges of the graph, respectively. 43 Chemical molecules can be represented as natural graphs, where atoms correspond to the nodes of the graph, and chemical bonds between atoms correspond to the edges. In traditional GNN models, only the feature information on nodes and the connectivity between nodes are considered during the computation process, without further utilizing other features of the edges.…”

Section: ■ Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Explainable Graph Neural Networks with Data Augmentation for Predicting pK_a of C–H Acids

An,

Liu,

Cai

et al. 2023

J. Chem. Inf. Model.

View full text Add to dashboard Cite

The pK a of C–H acids is an important parameter in the fields of organic synthesis, drug discovery, and materials science. However, the prediction of pK a is still a great challenge due to the limit of experimental data and the lack of chemical insight. Here, a new model for predicting the pK a values of C–H acids is proposed on the basis of graph neural networks (GNNs) and data augmentation. A message passing unit (MPU) was used to extract the topological and target-related information from the molecular graph data, and a readout layer was utilized to retrieve the information on the ionization site C atom. The retrieved information then was adopted to predict pK a by a fully connected network. Furthermore, to increase the diversity of the training data, a knowledge-infused data augmentation technique was established by replacing the H atoms in a molecule with substituents exhibiting different electronic effects. The MPU was pretrained with the augmented data. The efficacy of data augmentation was confirmed by visualizing the distribution of compounds with different substituents and by classifying compounds. The explainability of the model was studied by examining the change of pK a values when a specific atom was masked. This explainability was used to identify the key substituents for pK a. The model was evaluated on two data sets from the iBonD database. Dataset1 includes the experimental pK a values of C–H acids measured in DMSO, while dataset2 comprises the pK a values measured in water. The results show that the knowledge-infused data augmentation technique greatly improves the predictive accuracy of the model, especially when the number of samples is small.

show abstract

Section: Data Description and Methodsmentioning

confidence: 99%

Section: ■ Introductionmentioning

confidence: 99%

Explainable Graph Neural Networks with Data Augmentation for Predicting pK_a of C–H Acids

An,

Liu,

Cai

et al. 2023

J. Chem. Inf. Model.

View full text Add to dashboard Cite

show abstract

“…The actual embedding gradients g e are amalgamated with the pseudo item embedding gradients g p , and the unified gradient of the model g m and embeddings g e on the device is modified as: g = (g m , g e , g p ) (10) The second strategy involves incorporating Laplace noise into the feature embeddings processed through the GCN model to ensure local difference privacy. Subsequently, we clip the embeddings based on their L1-norm with a threshold δ, and introduce a LDP module with zero-mean Laplacian noise to the unified embeddings to fortify user privacy protection.…”

Section: The Privacy Protection Modulementioning

confidence: 99%

Differentially Private Knowledge Graph Neural Network for Recommeder System

胡,

刘

2023

Preprint

View full text Add to dashboard Cite

Knowledge Graphs (KGs) have been instrumental in mitigating the challenges of cold start and data scarcity in recommender systems, serving as a kind of auxiliary information. However, recent studies reveal that KGs, while beneficial, may also inadvertently furnish malicious attackers with more comprehensive background knowledge, escalating the risk of user privacy data breaches within these recommender systems. Current frameworks, which amalgamate KGs with privacy-preserving techniques, predominantly concentrate on enhancing privacy preservation, often at the expense of recommendation efficacy. In this research, we introduce a framework, named Differential Privacy Knowledge graph Neural network for Recommender systems (DPKNRec), designed to harmonize the trade-off between differential privacy noise and recommendation performance. Firstly, we formulate an integrated KG convolutional neural network adept at mining interrelationships between KG entities and item correlations effectively. Secondly, we incorporate a differential privacy technique to impose uniform Laplacian noise on feature embeddings and obfuscate genuine user-item interactions through pseudo-interaction item sampling, thereby augmenting user data security. Lastly, we evaluate our model using three distinct datasets-movies, music, and books. Empirical results from three benchmarks indicate that DPKNRec markedly supersedes contemporary benchmarks.

show abstract

“…It does not emphasise studies [ 17 , 18 , 19 ] that used graph-structured data from other medical sources, such as clinical data and longitudinal patient survey data. Meanwhile, Waikhom and Patgiri [ 20 ] reviewed the literature on using graph neural networks in various learning paradigms, including addressing the common formatting of graphical information and general standards or schemas that exist for the construction of graphical knowledge. However, no study in the present literature reviews disease prediction using graph ML approaches based on electronic health data.…”

Section: Introductionmentioning

confidence: 99%

Disease Prediction Using Graph Machine Learning Based on Electronic Health Data: A Review of Approaches and Trends

Uddin

2023

Healthcare

View full text Add to dashboard Cite

Graph machine-learning (ML) methods have recently attracted great attention and have made significant progress in graph applications. To date, most graph ML approaches have been evaluated on social networks, but they have not been comprehensively reviewed in the health informatics domain. Herein, a review of graph ML methods and their applications in the disease prediction domain based on electronic health data is presented in this study from two levels: node classification and link prediction. Commonly used graph ML approaches for these two levels are shallow embedding and graph neural networks (GNN). This study performs comprehensive research to identify articles that applied or proposed graph ML models on disease prediction using electronic health data. We considered journals and conferences from four digital library databases (i.e., PubMed, Scopus, ACM digital library, and IEEEXplore). Based on the identified articles, we review the present status of and trends in graph ML approaches for disease prediction using electronic health data. Even though GNN-based models have achieved outstanding results compared with the traditional ML methods in a wide range of disease prediction tasks, they still confront interpretability and dynamic graph challenges. Though the disease prediction field using ML techniques is still emerging, GNN-based models have the potential to be an excellent approach for disease prediction, which can be used in medical diagnosis, treatment, and the prognosis of diseases.

show abstract

A survey of graph neural networks in various learning paradigms: methods, applications, and challenges

Cited by 43 publications

References 216 publications

Explainable Graph Neural Networks with Data Augmentation for Predicting pK_a of C–H Acids

Explainable Graph Neural Networks with Data Augmentation for Predicting pK_a of C–H Acids

Differentially Private Knowledge Graph Neural Network for Recommeder System

Disease Prediction Using Graph Machine Learning Based on Electronic Health Data: A Review of Approaches and Trends

Contact Info

Product

Resources

About

A survey of graph neural networks in various learning paradigms: methods, applications, and challenges

Cited by 43 publications

References 216 publications

Explainable Graph Neural Networks with Data Augmentation for Predicting pKa of C–H Acids

Explainable Graph Neural Networks with Data Augmentation for Predicting pKa of C–H Acids

Differentially Private Knowledge Graph Neural Network for Recommeder System

Disease Prediction Using Graph Machine Learning Based on Electronic Health Data: A Review of Approaches and Trends

Contact Info

Product

Resources

About

Explainable Graph Neural Networks with Data Augmentation for Predicting pK_a of C–H Acids

Explainable Graph Neural Networks with Data Augmentation for Predicting pK_a of C–H Acids