Label-Aware Graph Convolutional Networks

Chen, Hao; Xu, Yue; Huang, Feiran; Deng, Zengde; Huang, Wenbing; Wang, Senzhang; He, Peng; Li, Zhoujun

doi:10.1145/3340531.3412139

Cited by 118 publications

(160 citation statements)

References 4 publications

Supporting

Mentioning

159

Contrasting

Order By: Relevance

“…One problem is that the representation of all nodes converges to the same value, along with repetitions of the graph convolution operation when the number of layers in the graph convolutional neural network is increased, resulting in reduced performance [ 29 , 30 , 31 ]. Another problem is that label information cannot be propagated to the entire graph when there is limited label information, resulting in low performance [ 32 , 33 , 34 ].…”

Section: Introductionmentioning

confidence: 99%

Prediction Models for Agonists and Antagonists of Molecular Initiation Events for Toxicity Pathways Using an Improved Deep-Learning-Based Quantitative Structure–Activity Relationship System

Matsuzaka

Totoki

Handa

et al. 2021

IJMS

View full text Add to dashboard Cite

In silico approaches have been studied intensively to assess the toxicological risk of various chemical compounds as alternatives to traditional in vivo animal tests. Among these approaches, quantitative structure–activity relationship (QSAR) analysis has the advantages that it is able to construct models to predict the biological properties of chemicals based on structural information. Previously, we reported a deep learning (DL) algorithm-based QSAR approach called DeepSnap-DL for high-performance prediction modeling of the agonist and antagonist activity of key molecules in molecular initiating events in toxicological pathways using optimized hyperparameters. In the present study, to achieve high throughput in the DeepSnap-DL system–which consists of the preparation of three-dimensional molecular structures of chemical compounds, the generation of snapshot images from the three-dimensional chemical structures, DL, and statistical calculations—we propose an improved DeepSnap-DL approach. Using this improved system, we constructed 59 prediction models for the agonist and antagonist activity of key molecules in the Tox21 10K library. The results indicate that modeling of the agonist and antagonist activity with high prediction performance and high throughput can be achieved by optimizing suitable parameters in the improved DeepSnap-DL system.

show abstract

Section: Introductionmentioning

confidence: 99%

Prediction Models for Agonists and Antagonists of Molecular Initiation Events for Toxicity Pathways Using an Improved Deep-Learning-Based Quantitative Structure–Activity Relationship System

Matsuzaka

Totoki

Handa

et al. 2021

IJMS

View full text Add to dashboard Cite

show abstract

“…We apply the SAGNN on datasets Twitter15 and Twitter16 to evaluate its performance. A GCNII (Graph Convolutional Network via Initial residual and Identity mapping) network [21] serves as the baseline to assess the proposed algorithm. In this experiment, the SAGNN has two aggregation layers, unlike the one shown in Figure 4, in which only a single aggregation layer is shown for clarity.…”

Section: Network Setupmentioning

confidence: 99%

Rumor Detection Based on SAGNN: Simplified Aggregation Graph Neural Networks

Zhang

Li²,

Zhou

et al. 2021

MAKE

View full text Add to dashboard Cite

Identifying fake news on media has been an important issue. This is especially true considering the wide spread of rumors on popular social networks such as Twitter. Various kinds of techniques have been proposed for automatic rumor detection. In this work, we study the application of graph neural networks for rumor classification at a lower level, instead of applying existing neural network architectures to detect rumors. The responses to true rumors and false rumors display distinct characteristics. This suggests that it is essential to capture such interactions in an effective manner for a deep learning network to achieve better rumor detection performance. To this end we present a simplified aggregation graph neural network architecture. Experiments on publicly available Twitter datasets demonstrate that the proposed network has performance on a par with or even better than that of state-of-the-art graph convolutional networks, while significantly reducing the computational complexity.

show abstract

“…We improve the GCN layer of graph U-Net to GCNII layer [31], the formula of forward propagation is defined as:…”

Section: Improved Information Aggregation Layer Bgnnmentioning

confidence: 99%

“…In order to ensure the back-propagation and gradient descent algorithm works, we not only change the original skip-connection way of graph U-Net from residual to dense, but also add a deep supervision mechanism to the loss function with four outputs of different encoders. The graph U-Net+ extends the information aggregation approach by combining GCNII [31] and bilinear information aggregator [32] for all encoding and decoding blocks. In addition, we also introduce a novel graph normalization technique named NodeNorm [33], so as to eliminate the over-smoothing problem with Fig.…”

Section: Introductionmentioning

confidence: 99%

Social behavior prediction with graph U-Net+

Yan

Cao

2021

Discov Internet Things

View full text Add to dashboard Cite

We focus on the problem of predicting social media user’s future behavior and consider it as a graph node binary classification task. Existing works use graph representation learning methods to give each node an embedding vector, then update the node representations by designing different information passing and aggregation mechanisms, like GCN or GAT methods. In this paper, we follow the fact that social media users have influence on their neighbor area, and extract subgraph structures from real-world social networks. We propose an encoder–decoder architecture based on graph U-Net, known as the graph U-Net+. In order to improve the feature extraction capability in convolutional process and eliminate the effect of over-smoothing problem, we introduce the bilinear information aggregator and NodeNorm normalization approaches into both encoding and decoding blocks. We reuse four datasets from DeepInf and extensive experimental results demonstrate that our methods achieve better performance than previous models.

show abstract

Label-Aware Graph Convolutional Networks

Cited by 118 publications

References 4 publications

Prediction Models for Agonists and Antagonists of Molecular Initiation Events for Toxicity Pathways Using an Improved Deep-Learning-Based Quantitative Structure–Activity Relationship System

Prediction Models for Agonists and Antagonists of Molecular Initiation Events for Toxicity Pathways Using an Improved Deep-Learning-Based Quantitative Structure–Activity Relationship System

Rumor Detection Based on SAGNN: Simplified Aggregation Graph Neural Networks

Social behavior prediction with graph U-Net+

Contact Info

Product

Resources

About