# Generating a matching model from a Grouping campaign

This scenario applies only to subscription-based Talend Platform products with Big Data and Talend Data Fabric.

tMatchModel reads the sample of suspect pairs computed on a list of duplicate childhood education centers and labeled by data stewards in Talend Data Stewardship. It generates several matching models, searches the best combination of the learning parameters and keeps the best matching model which comes out as the result of cross validation.

Prerequisites:

* You have generated the suspect data pairs by using the tMatchPairing component and labeled them in Talend Data Stewardship. For further information, see [Computing suspect pairs and writing a sample in Talend Data Stewardship](https://help.qlik.com/talend/en-US/data-matching/8.0/computing-suspect-pairs-writing-sample-in-stewardship).

  For further information about handling grouping tasks to decide on relationship among pairs of records, see [Talend Data Stewardship Examples](https://help.qlik.com/talend/en-US/data-stewardship-user-guide/8.0/talend-data-stewardship-examples).

### In this section

* [Setting up the Job](https://help.qlik.com/talend/en-US/data-matching/8.0/generating-matching-model-grouping-setting-up-job)
* [Selecting the Spark mode](https://help.qlik.com/talend/en-US/data-matching/8.0/selecting-spark-mode-spark-batch?id=97)
* [Configuring the connection to the file system to be used by Spark](https://help.qlik.com/talend/en-US/data-matching/8.0/configuring-connection-to-file-system-to-be-used-by-spark?id=98)
* [Generating the matching model](https://help.qlik.com/talend/en-US/data-matching/8.0/generating-matching-model-grouping-generating-matching-model)