Recommendations with Elastic Graph

Lightweight recommendations with Elasticsearch

Welcome! This demo by OpenSource Connections demonstrates the power of Elasticsearch's Graph plugin to build and tune recommender systems.

Start by adding a few movies that you like below:

x

Elastic Graph uses search to find users most similar to you, finding movies you've missed but folks just like you consider significant.

Play with user matchmaking settings, and see what you think of the results

Quite a lot of Relevance Work can be brought to bear to improve Elastic Graph's ability to find similar users. Below, you can match directly to specific attributes of preferred movies:

Use preferred movie's:

   

What do you think of the results? What other data about you/users could improve how easily we find users similar to you?

We can do so much more with Elastic Graph. Let's reach beyond direct recommendations to see how your recommendations are linked to others. We can guide globally, or scope our graph-based guidance to users similar to you.

Scoped to:

Considering?

You might also like:

How else can Elastic Graph improve the relevance in your search/discovery apps?

  • Personalized Search -- use Elastic Graph's rankings to reweight search results for a user.
  • Social Search -- can we reweight search according to direct social relationships?
  • Taste Profiles -- bucketing recommendations and search results by specific tastes, genres, or other notable properties
  • Learn more about OpenSource Connections relevance services.