2 Graph Embeddings
This chapter covers
- Graph embeddings and their importance
- Creating node embeddings using non-GNN and GNN methods
- Comparing node embeddings on a semi-supervised problem
- A deeper dive into embedding methods
In this chapter, we explore graph embeddings, an essential tool in graph-based machine learning. Graph embeddings transform the intricate structure of graphs—be it the entire graph, individual nodes, or edges—into a more manageable, lower-dimensional space. We do this to compress a complex dataset into a form that's easier to work with, without losing its inherent patterns and relationships, the information to which we will apply a GNN or other machine learning method.
Graphs, as we have learned, encapsulate relationships and interactions within a network, whether it's social networks, biological networks, or any system where entities are interconnected. Embeddings capture these real life relationships in a compact form, facilitating tasks like visualization, clustering, or predictive modeling.
There are numerous strategies to derive these embeddings, each with its unique approach and application. From classical graph algorithms that leverage the network's topology, to linear algebra techniques that decompose matrices representing the graph, and onto more advanced methods like GNNs. GNNs stand out because they can integrate the embedding process directly into the learning algorithm itself.