GR-GNN: Gated Recursion-Based Graph Neural Network Algorithm
Under an internet background involving artificial intelligence and big data—unstructured, materialized, network graph-structured data, such as social networks, knowledge graphs, and compound molecules, have gradually entered into various specific business scenarios. One problem that urgently needs t...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-04-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/10/7/1171 |
| _version_ | 1827623099819884544 |
|---|---|
| author | Kao Ge Jian-Qiang Zhao Yan-Yong Zhao |
| author_facet | Kao Ge Jian-Qiang Zhao Yan-Yong Zhao |
| author_sort | Kao Ge |
| collection | DOAJ |
| description | Under an internet background involving artificial intelligence and big data—unstructured, materialized, network graph-structured data, such as social networks, knowledge graphs, and compound molecules, have gradually entered into various specific business scenarios. One problem that urgently needs to be solved in the industry involves how to perform feature extractions, transformations, and operations in graph-structured data to solve downstream tasks, such as node classifications and graph classifications in actual business scenarios. Therefore, this paper proposes a gated recursion-based graph neural network (GR-GNN) algorithm to solve tasks such as node depth-dependent feature extractions and node classifications for graph-structured data. The GRU neural network unit was used to complete the node classification task and, thereby, construct the GR-GNN model. In order to verify the accuracy, effectiveness, and superiority of the algorithm on the open datasets Cora, CiteseerX, and PubMed, the algorithm was used to compare the operation results with the classical graph neural network baseline algorithms GCN, GAT, and GraphSAGE, respectively. The experimental results show that, on the validation set, the accuracy and target loss of the GR-GNN algorithm are better than or equal to other baseline algorithms; in terms of algorithm convergence speed, the performance of the GR-GNN algorithm is comparable to that of the GCN algorithm, which is higher than other algorithms. The research results show that the GR-GNN algorithm proposed in this paper has high accuracy and computational efficiency, and very wide application significance. |
| first_indexed | 2024-03-09T11:38:40Z |
| format | Article |
| id | doaj.art-e26d7d9c10ac4cfc9ae87f6c6968d6da |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-09T11:38:40Z |
| publishDate | 2022-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-e26d7d9c10ac4cfc9ae87f6c6968d6da2023-11-30T23:38:12ZengMDPI AGMathematics2227-73902022-04-01107117110.3390/math10071171GR-GNN: Gated Recursion-Based Graph Neural Network AlgorithmKao Ge0Jian-Qiang Zhao1Yan-Yong Zhao2Nanjing Institute of Software Technology, Institute of Software, Chinese Academy of Sciences, Nanjing 211135, ChinaSchool of Mathematics and Statistics, Xuzhou University of Technology, Xuzhou 221018, ChinaDepartment of Statistics, Nanjing Audit University, Nanjing 211815, ChinaUnder an internet background involving artificial intelligence and big data—unstructured, materialized, network graph-structured data, such as social networks, knowledge graphs, and compound molecules, have gradually entered into various specific business scenarios. One problem that urgently needs to be solved in the industry involves how to perform feature extractions, transformations, and operations in graph-structured data to solve downstream tasks, such as node classifications and graph classifications in actual business scenarios. Therefore, this paper proposes a gated recursion-based graph neural network (GR-GNN) algorithm to solve tasks such as node depth-dependent feature extractions and node classifications for graph-structured data. The GRU neural network unit was used to complete the node classification task and, thereby, construct the GR-GNN model. In order to verify the accuracy, effectiveness, and superiority of the algorithm on the open datasets Cora, CiteseerX, and PubMed, the algorithm was used to compare the operation results with the classical graph neural network baseline algorithms GCN, GAT, and GraphSAGE, respectively. The experimental results show that, on the validation set, the accuracy and target loss of the GR-GNN algorithm are better than or equal to other baseline algorithms; in terms of algorithm convergence speed, the performance of the GR-GNN algorithm is comparable to that of the GCN algorithm, which is higher than other algorithms. The research results show that the GR-GNN algorithm proposed in this paper has high accuracy and computational efficiency, and very wide application significance.https://www.mdpi.com/2227-7390/10/7/1171GR-GNNgraph neural networkbias random walksGRU |
| spellingShingle | Kao Ge Jian-Qiang Zhao Yan-Yong Zhao GR-GNN: Gated Recursion-Based Graph Neural Network Algorithm Mathematics GR-GNN graph neural network bias random walks GRU |
| title | GR-GNN: Gated Recursion-Based Graph Neural Network Algorithm |
| title_full | GR-GNN: Gated Recursion-Based Graph Neural Network Algorithm |
| title_fullStr | GR-GNN: Gated Recursion-Based Graph Neural Network Algorithm |
| title_full_unstemmed | GR-GNN: Gated Recursion-Based Graph Neural Network Algorithm |
| title_short | GR-GNN: Gated Recursion-Based Graph Neural Network Algorithm |
| title_sort | gr gnn gated recursion based graph neural network algorithm |
| topic | GR-GNN graph neural network bias random walks GRU |
| url | https://www.mdpi.com/2227-7390/10/7/1171 |
| work_keys_str_mv | AT kaoge grgnngatedrecursionbasedgraphneuralnetworkalgorithm AT jianqiangzhao grgnngatedrecursionbasedgraphneuralnetworkalgorithm AT yanyongzhao grgnngatedrecursionbasedgraphneuralnetworkalgorithm |