Whether you're trying to recommend products to buy, people to follow, influencers to target or articles to read, recommendations using graph algorithms are very valuable to produce.

Typical workflows involve using ML to train a model, and using that model to power recommendations. This requires the models be re-trained as new data comes in. This makes it challenging to ensure you're serving relevant recommendations based on fast-changing data across many systems.

This session will cover how to combine graph analytics queries, which are often global and interactive, with local graph traversals to generate real-time recommendations. OLAP + OLTP ftw!