How to Enable Result Clustering in Fusion

Result clustering is a feature that automatically finds groups of results related to the current search query.  You can read more about this in the Solr documentation.  Here's how to configure it in Fusion:

  1. Create a collection in the Fusion Admin UI. I named mine “cluster”.

 

 

2. Now go into the Query Pipeline for the “cluster” collection and set up the query parameters for your desired clustering implementation. This is where you’ll set the params that you’d normally set in the /clustering request handler of the Solrconfig. These parameters will be specific to your configuration.

Refer to https://wiki.apache.org/solr/ClusteringComponent for more information on setting parameters and using the clustering component in general. 

 

 

3. In your “cluster” collection, go to the Solr Config and solrconfig.xml. Here, you’ll need to modify the lib directories since Fusion does not set ${solr.install.dir...} to find the contrib and dist folders.

 

 

 

4. Instead of using the /clustering request handler, you can use /select and add “clustering” as a last component. Since you’ve already set up the clustering parameters in the query pipeline, this will work just fine.

 

 

 

5. Now enable the clustering component by changing enabled=“${solr.clustering.enabled:false}” to enabled=“${solr.clustering.enabled:true}” and save the changes to your configuration. (You could also just omit the line completely and it will default to true)

 

6. Finally, to ensure that clustering has been set up correctly, visit your collection in the Solr admin using the /select handler and the clustering parameters clustering=true, clustering.collection=true, and results=true and the “clusters” array should show up as the final component in your results.

 

Prepared By: 

Esther Quansah

Have more questions? Submit a request

0 Comments

Please sign in to leave a comment.
Powered by Zendesk