1st Special Session on Graph Neural Networks for Real World Data 

within

2024 IEEE International Conference on Evolving and Adaptive Intelligent Systems (EAIS)

Madrid, Spain

23-24 May 2024

Submission deadline: 20 February2024

Graphs are a powerful tool for the analysis and depiction of real-world data due to their ability to capture intricate relationships and structures inherent in many domains. These structures are often graph-like, and the interaction among various sources of information can be effectively represented using nodes and links.

The application of the computational versatility of neural architectures to graphs has resulted in the development of powerful computational models, notably Graph Neural Networks (GNN). These models excel in exploiting the inherent information present in a graph structure. GNNs offer a versatile framework for various computations, including but not limited to node classification, graph classification, and link prediction. These computations play a crucial role in solving both supervised and unsupervised classification problems.

Moreover, GNNs prove their adaptability by accommodating information that doesn't conform to traditional grid-like structures. This flexibility allows data to be cast into graphs, considering a topology derived from specific features and adjacency matrices. By doing so, GNNs become applicable to a broader range of data types and structures.

The overarching goal of this proposed special session is to provide a forum for the presentation and discussion of original papers and reviews on the latest methods involving GNNs. These methods are specifically tailored for the analysis of diverse real-world data types. Examples of such data include environmental data, biomedical data, and social network data. By exploring the application of GNNs across different domains, we aim to foster a deeper understanding of their capabilities and potential contributions to advancing data analysis and representation in various fields.

Relevant topics within this context include:

• General GNN-based architectures

• Representation Learning by GNNs

• Explainability in GNNs

• GNNs in Computer vision and Image Processing

• GNNs in Natural Language Processing

• Processing of Biological data with GNNs

• GNNs for Environmental Monitoring

Researchers from academia, industry and government organizations are invited to submit original papers within the technical scope of IEEE EAIS 2024. Authors of selected papers will be invited to submit extended versions for possible inclusion in a special issue of the Journal of Evolving Systems (Springer)