How do diseases and information spread? How can we predict traffic or weather? Answering these questions requires massive amounts of data. Complex data can be represented as a graph of relationships and interactions between objects. Graph data structures can be ingested by algorithms such as neural networks to perform tasks including classification, clustering, and regression.

This course explores the computational, algorithmic, and modeling challenges specific to the analysis of massive graphs. By studying underlying graph structures, you will master machine learning and data mining techniques that can improve prediction and reveal insights on a variety of networks.

  • Build more accurate machine learning models by understanding the underlying relational structures of your data.
  • Understand and apply traditional methods for machine learning on graphs, such as node embeddings and PageRank.
  • Leverage graph-structured data and make better predictions using graph neural networks. Construct your own graph neural network using PyTorch Geometric.
  • Expand your understanding of data by incorporating different node and edge types in knowledge graphs.
  • Discover recurring and significant patterns of interconnections in your data with network motifs and community structure.
  • Scale up your neural networks with generative models for graphs.