Graph Algorithms and Graph Machine Learning: Making Sense of Today's Choices

Graph machine learning is coming of age, with rapid research development and growing interest. Users now have several alternatives to choose from: graph pattern matching, graph algorithms, graph embeddings, and graph neutral networks.

Does graph machine learning replace graph algorithms? How does graph machine learning different from other machine learning? What sort of system do I need to run these analytical techniques?

This talk provides a description of these four categories of graph analytics, what sort of problems they tackle, their benefits, and their requirements.

The explanations will be illustrated with examples for fraud detection, recommendation, supply chain management, and other real-world use cases.

Talk+Live Q&A at the Eastern Auditorium in Connected Data World Center; Free Live Streaming

You need an access pass to attend this session: Full Access Pass, Diversity Access Pass or Free Access Pass apply