Abstract: 

Relationships are highly predictive of behavior, yet most data science models overlook this information as it's difficult to extract network structure to use at scale in machine learning (ML).

With graphs, relationships are embedded in the data itself, making it practical to add these predictive capabilities to your existing practices.

In this session, you’ll learn more about:

Using graph-native ML to make break-through predictions
Taking different approaches to graph feature engineering from queries and algorithms to embeddings
How Neo4j has democratized graph-based ML techniques, leveraging everything from classical network science approaches to deep learning and graph convolutional neural networks

We’ll also walk through how to generate representations of your graph using graph embeddings, create ML models for link prediction or node classification, and apply these models to add missing information to an existing graph or incoming graph data.

Bio: 

Alicia Frame is the lead product manager for data science at Neo4j. She's spent the last year translating input from customers, early adopters, and the community into the first truly enterprise product for doing data science with graphs: Neo4j's Graph Data Science Library. She has a Ph.D. in computational biology from UNC Chapel Hill, and her background is in data science applications in healthcare and life sciences.

She's worked in academia, government, and the private sector to leverage graph techniques for drug discovery, molecular optimization, and risk assessments -- and is super excited to be making it possible for anyone to use advanced graph techniques with Neo4j.