Supervisr: Matching Rules

Manage objects which are accessed through the DataGroomr platform and manage your organization algorithmic models used to identify duplicate records.

Matching Rules: Accessing
Matching Rules can be accessed directly by pressing the Matching Rules link located under the Supervisr tab in the Navigation Menu.Then select the the appropriate object. All the existing rules for the selected object will be displayed.
Wed, 25 Nov, 2020 at 2:03 PM
Matching Rules: Creating
Matching Rules are based on models powered by machine learning.  There are two ways to create a new model.  You can press the ADD MODEL button to develop one from scratch or you can select an existing rule and then press the CLONE button.  In either case, the Model configuration window will be opened which includes...
Wed, 23 Jun, 2021 at 2:33 PM
Matching Rules: Creating/Editing Fields
For new rules or those in Draft status users may edit the matching (or comparison) type used for each field.  This is done by clicking on the pencil icon inside a field which brings up a drop down dialogue as shown below. The prefilled matching type is auto selected by DataGroomr based on the type of objec...
Wed, 25 Nov, 2020 at 5:38 PM
Matching Rules: Training
Matching Rules are based on models powered by machine learning.  When a new model is created it will need to be Trained, which is a process where Datagroomr will look for duplicates based on the fields you specified.  But in order for Machine Learning models to work, a person must confirm (or reject) a small subset...
Sun, 13 Dec, 2020 at 1:51 PM
Matching Rules: Editing
To Edit an existing matching rule, select it from the list and then press the EDIT button.  If the rule is in Draft status, all the entry fields will be editable.  You can add a name, add/remove/edit fields and enable auto training. Tip: To learn more about editing fields, please follow this link. For new o...
Sun, 13 Dec, 2020 at 1:56 PM
Machine Learning Models: Cloning
Occasionally you may need to modify or retrain an existing matching rule.  For example, you may need to remove or add a field.  However an existing rule cannot be changed this way, but you can create a copy that can be edited. To do this, select the rule and then press the CLONE button. 
Sun, 13 Dec, 2020 at 2:05 PM
Matching Rules: Re-training
Occasionally an existing rule needs to be retrained.  This sometimes happens when a rule is not detecting all the duplicates or a large amount of new data has been added to Salesforce.  Notice that an existing rule has a version number.  To re-train this rule, select it from the list and press OPEN.   The...
Sun, 13 Dec, 2020 at 2:14 PM
Matching Rules: Assign (to Dataset)
Once a model is trained it becomes a matching rule that is available to be used by a dataset.  There are a couple ways to do that.   Option 1:   From the Matching Rules feature, select a rule press the Assign.  Then choose the datasets that apply and the Assign button. Option 2: Alternatively, the same...
Sun, 13 Dec, 2020 at 2:24 PM
Machine Learning Models: Setting a Default
A model that is marked as as Default will be used by TRIMMR dataset of the same object type automatically.  For objects with multiple models, users can select any model in the dataset configurations (see Configure Datasets documentation for more info). To set a model as a default, select the model and then pres...
Sun, 13 Dec, 2020 at 2:01 PM
Machine Learning Models: Deleting
A model can be deleted by selecting it and then pressing the Trash button.  Good to know: Rules that are used by a dataset cannot be deleted.  
Wed, 25 Nov, 2020 at 7:53 PM