When it comes to recommendation systems, embeddings have taken the Natural Language Processing ML world by storm but they are also enabling new services, based on finding similar items, where models can be trained without any labelled data.

In this webinar, we will talk about real-time personalization in search ranking and similar listing recommendations. We will describe how to build and operate search and recommendation systems on Hopsworks, with support for the most important pieces of infrastructure: an embeddings store for retrieving matching items and the Hopsworks Feature Store to enrich results to enable them to be ranked by ML models.‍

‍

What you will learn:

• What are Recommender Systems and Personalized Search?
• What are Embeddings and how do we create them?
• How to integrate Embedding Stores and Feature Stores.

‍

We will also have a live demo on how to use Embeddings and the Hopsworks Feature Store and Elasticsearch.