Configure Connect Component algorithm and Levenshtein Distance method for a clustering solution

Apply Configure Connect Component and Levenshtein Distance method encoding to optimize the training for your clustering solutions.

始める前に

Role required: admin or ml_admin

注:

Configuring advanced settings on your ML solutions is optional. If you choose to configure any of these settings, make sure you're well informed regarding the technology you're enabling in the solution, and that your use case benefits from what the technology offers. For more information, see the Dive deeper with Clustering Advanced Parameters article on ServiceNow Community.

• Create and train a clustering solution definition or use an existing one.
• Role required: admin or ml_admin

このタスクについて

When training clustering solutions, you have the following three options.

• Use the default k-means algorithm.
• Use the optional DBSCAN solution parameter with the Euclidean distance method as a metric.
• Use the optional DBSCAN, Minimum Neighbors, and Levenshtein Distance solution parameters. Connect Component is enabled by DBSCAN and Minimum Neighbors, and supports both Paragraph Vector-based text and Levenshtein Distance-based text. If you train your solution using the Levenshtein Distance method, you don't need to use a word corpus in your clustering solution.

In this example scenario, you train your solution definition by using the third option referenced above.

手順

1. Navigate to All > Predictive Intelligence > Clustering > Solution Definitions.
2. Open a trained clustering solution definition form.
3. On the Advanced Solution Settings tab in the Related Links section of the form, select New.

   

4. Create a parameter record.
   1. In the Solution Parameters field, select the search icon.
   2. In the ML Solution Parameters screen, select Levenshtein Distance.

   

5. Select Submit.

   The Advanced Solution Setting record screen refreshes.

   

6. Select Submit.

   Result: Levenshtein Distance is configured for your clustering solution. Its solution parameter appears on the Advanced Solution Settings tab of your clustering definition form.

   

7. Repeat steps 1-6 from the previous Levenshtein Distance example, except this time you're creating the Minimum Neighbors and DBSCAN solution parameters, which together enable the Connect Component feature.

   

   When you select, configure, and submit the Minimum Neighbors solution parameter, be sure to set the User Inputs field with a value of 1. Only some parameters have a User Inputs field.

   

   Result:

   Connect Component is configured for your clustering solution. Its two solution parameters appear on the Advanced Solution Settings tab of your clustering definition form, alongside the Levenshtein Distance parameter you configured in steps 1-6 of this procedure.

   

関連タスク

• Create and train a clustering solution