Graph Neural Networks Using Local Descriptions in Attributed Graphs: An Application to Symbol Recognition and Hand Written Character Recognition

Kajla, Nadeem Iqbal; Missen, Malik Muhammad Saad; Luqman, Muhammad; Coustaty, Mickaël

doi:10.1109/access.2021.3096845

Cited by 14 publications

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…P a t h s S t r a t i f i e d P a t h s C l i q u e s S t r a t i f i e d C l i q u e s In Table 3, we finally show a brief comparison against current approaches for graph classification. Competitors span a variety of techniques, including classifiers working on the top of GEDs [49,67], kernel methods [68][69][70][71] and several embedding techniques [68,72,73], including Granular Computing-based [32,74] and those based on neural networks and deep learning [75][76][77][78][79]. We can see that our method has comparable performances against current approaches in the graph classification literature.…”

Section: Computational Resultsmentioning

confidence: 80%

On Information Granulation via Data Clustering for Granular Computing-Based Pattern Recognition: A Graph Embedding Case Study

2022

View full text Add to dashboard Cite

Granular Computing is a powerful information processing paradigm, particularly useful for the synthesis of pattern recognition systems in structured domains (e.g., graphs or sequences). According to this paradigm, granules of information play the pivotal role of describing the underlying (possibly complex) process, starting from the available data. Under a pattern recognition viewpoint, granules of information can be exploited for the synthesis of semantically sound embedding spaces, where common supervised or unsupervised problems can be solved via standard machine learning algorithms. In this work, we show a comparison between different strategies for the automatic synthesis of information granules in the context of graph classification. These strategies mainly differ on the specific topology adopted for subgraphs considered as candidate information granules and the possibility of using or neglecting the ground-truth class labels in the granulation process. Computational results on 10 different open-access datasets show that by using a class-aware granulation, performances tend to improve (regardless of the information granules topology), counterbalanced by a possibly higher number of information granules.

show abstract

Section: Computational Resultsmentioning

confidence: 80%

On Information Granulation via Data Clustering for Granular Computing-Based Pattern Recognition: A Graph Embedding Case Study

2022

View full text Add to dashboard Cite

show abstract

“…represents the number of master information nodes after aggregation, d represents the initial embedded dimension, and the generated dynamic adjacency matrix is defined as: = ∈ 3×3 (13) A local graph convolution method is proposed in document [33] to learn the information and distance of the graph and describe the graph locally through the node and edge information of the graph. In this paper, the graph convolution is defined in the vertex domain, omitting the edge information, so it is not necessary to calculate the Laplace determinant of the graph, and each convolution is a global description of the graph.…”

Section: B Information Aggregation Modulementioning

confidence: 99%

Traffic Flow Prediction Based on Information Aggregation and Comprehensive Temporal-Spatial Synchronous Graph Neural Network

Cheng

Zhang

et al. 2023

IEEE Access

View full text Add to dashboard Cite

It is extremely important to build a reasonable traffic network structure for traffic flow prediction. Owing to the complexity and dynamic of traffic networks, the graph neural network model has become one of the most effective methods for mining the spatial-temporal relationship between traffic flow data. However, most current methods use two components to extract the spatial dependence and time dependence separately and do not consider the auxiliary effect of additional traffic factors on the prediction target. Based on the above problems, this paper proposes a neural network prediction model for a comprehensive spatial-temporal synchronous graph based on information aggregation. The model is composed of a fusion feature attention module, an information aggregation module, and a comprehensive information integration framework. The fusion feature attention module considers the impact of each traffic factor on the traffic flow and strengthens the internal relationship of various traffic features; the information aggregation module synchronously extracts the temporal-spatial dependence of traffic flow; The multiinformation combination module combines the traffic flow with the secondary information to mine the hidden relationship between the primary and secondary information. The experimental results on two realworld datasets show that the prediction effect of the model set out in the present paper is significantly better than that of the baseline.

show abstract

“…GAT and MoNet extend GCN by leveraging an explicit attention mechanism ( Lee et al, 2019 ). Due to powerful represent capabilities, GNNs have been applied into a wide range of applications including knowledge graphs ( Zhang, Cui & Zhu, 2020 ), molecular graph generation ( De Cao & Kipf, 2018 ), graph metric learning and image recognition ( Kajla et al, 2021 ; Riba et al, 2021 ). Recently, graph sampling was investigated in GNNs for scaling to larger graphs and better generalization.…”

Section: Related Workmentioning

confidence: 99%

Evaluating graph neural networks under graph sampling scenarios

Wei

2022

PeerJ Computer Science

View full text Add to dashboard Cite

Background It is often the case that only a portion of the underlying network structure is observed in real-world settings. However, as most network analysis methods are built on a complete network structure, the natural questions to ask are: (a) how well these methods perform with incomplete network structure, (b) which structural observation and network analysis method to choose for a specific task, and (c) is it beneficial to complete the missing structure. Methods In this paper, we consider the incomplete network structure as one random sampling instance from a complete graph, and we choose graph neural networks (GNNs), which have achieved promising results on various graph learning tasks, as the representative of network analysis methods. To identify the robustness of GNNs under graph sampling scenarios, we systemically evaluated six state-of-the-art GNNs under four commonly used graph sampling methods. Results We show that GNNs can still be applied on single static networks under graph sampling scenarios, and simpler GNN models are able to outperform more sophisticated ones in a fairly experimental procedure. More importantly, we find that completing the sampled subgraph does improve the performance of downstream tasks in most cases; however, completion is not always effective and needs to be evaluated for a specific dataset. Our code is available at https://github.com/weiqianglg/evaluate-GNNs-under-graph-sampling.

show abstract

Graph Neural Networks Using Local Descriptions in Attributed Graphs: An Application to Symbol Recognition and Hand Written Character Recognition

Cited by 14 publications

References 25 publications

On Information Granulation via Data Clustering for Granular Computing-Based Pattern Recognition: A Graph Embedding Case Study

On Information Granulation via Data Clustering for Granular Computing-Based Pattern Recognition: A Graph Embedding Case Study

Traffic Flow Prediction Based on Information Aggregation and Comprehensive Temporal-Spatial Synchronous Graph Neural Network

Evaluating graph neural networks under graph sampling scenarios

Contact Info

Product

Resources

About