Join us to understand how you can use graph-native machine learning in Neo4j to make break-through predictions. Previously only accessible to researchers and a very few advanced tech companies, Neo4j has democratized graph-based ML techniques that leverage deep learning and graph convolutional neural networks.
Most data science models ignore network structure, while graphs add highly predictive features to ML models, increasing accuracy and enabling otherwise unattainable predictions based on relationships. With the recent update to the Neo4j Graph Data Science library, anyone can take advantage of this state-of-the-science technique to create representations of your graph’s most significant features for new and more accurate predictions with the data you already have.
In this session, we’ll explain our new graph embeddings and demonstrate using the GraphSAGE embedding results with our new ML catalog. We’ll also visualize the predictions of different models using Neo4j Bloom.
• Alicia Frame, Lead Product Manager for Graph Data Science, Neo4j at Neo4j
• Amy Hodler, Director, Neo4j Graph Analytics & AI Programs at Neo4j