Social network analysis
Social networks provide rich relational data that can be naturally represented as graphs, with users as nodes and connections between users as edges. Graph deep learning techniques have emerged as powerful tools for analyzing and extracting insights from social network data.
Community detection
Community detection aims to identify clusters or groups of densely connected users within a social network, as shown in Figure 11.1. Traditional community detection algorithms, such as modularity optimization or spectral clustering, often struggle with large-scale networks. GNNs offer a promising alternative by learning node embeddings that capture both local and global network structures:
Figure 11.1 – Community detection
For instance, Wang et al. (2024) (https://doi.org/10.1016/j.neucom.2024.127703) proposed a graph autoencoder (GAE) approach for community detection. The model uses graph convolutional layers to encode nodes...