Elasticsearch Learning to Rank: the documentation¶
Learning to Rank applies machine learning to relevance ranking. The Elasticsearch Learning to Rank plugin (Elasticsearch LTR) gives you tools to train and use ranking models in Elasticsearch. This plugin powers search at places like Wikimedia Foundation and Snagajob.
- Want a quickstart? Check out the demo in hello-ltr.
- Brand new to learning to rank? head to Core Concepts.
- Otherwise, start with How does the plugin fit in?
Pre-built versions can be found here. Want a build for an ES version? Follow the instructions in the README for building or create an issue. Once you’ve found a version compatible with your Elasticsearch, you’d run a command such as:
./bin/elasticsearch-plugin install \ https://github.com/o19s/elasticsearch-learning-to-rank/releases/download/v1.5.4-es7.11.2/ltr-plugin-v1.5.4-es7.11.2.zip
(It’s expected you’ll confirm some security exceptions, you can pass -b to elasticsearch-plugin to automatically install)
Are you using x-pack security in your cluster? we got you covered, check On XPack Support (Security) for specific configuration details.
The plugin and guide was built by the search relevance consultants at OpenSource Connections in partnership with the Wikimedia Foundation and Snagajob Engineering. Please contact OpenSource Connections or create an issue if you have any questions or feedback.
- Core Concepts
- How does the plugin fit in?
- Working with Features
- What is a feature in Elasticsearch LTR?
- Features are Mustache Templated Elasticsearch Queries
- Uploading and Naming Features
- Initialize the default feature store
- Features and feature sets
- Create a feature set
- Feature set CRUD
- Validating features
- Adding to an existing feature set
- Feature Names are Unique
- Feature Sets are Lists
- But wait there’s more
- Feature Engineering
- Logging Feature Scores
- Uploading A Trained Model
- Searching with LTR
- On XPack Support (Security)
- Advanced Functionality