Migrating enrichments from Watson Explorer

Types of resources that can be migrated

The following types of resources can be migrated from Watson Explorer to Discovery:

• From Watson Explorer Analytical Components: User dictionaries
• From Watson Explorer oneWEX: Dictionaries, character patterns, and facets.

To analyze data with these migrated enrichments, you can use a Content Mining project. The tools in the associated Content Mining application are similar to tools that are available in Watson Explorer.

• For more information about how to create a Content Mining project, see Creating projects.
• For more information about how to apply enrichments to a collection in the Content Mining application, see Applying the annotator.

Importing dictionaries from Watson Explorer Analytical Components

You can import user dictionaries from IBM Watson Explorer Analytical Components.

The default file location and name for dictionaries that are saved in Watson Explorer Analytical Components is ${primary_server_node}/{primary_configuration}/{collection_ID}/{dictionary_name}.fdic.xml.

1. Download your user dictionaries from Watson Explorer Analytical Components.

2. From your Discovery Content mining project, open the Content Mining application.

3. From the analysis view of your collection, click the Collections link in the breadcrumb to open the Create a collection for your analytics solutions page of the Content Mining application.

4. To create an annotator, click collection, and then select custom annotator from the list.

   

5. Click Create custom annotator.

6. Name your annotator, and then optionally add a description.

7. From the Annotator Type menu, select Dictionary, and then click Next.

8. Click the Import button, and then select the {name}.fdic.xml dictionary file that you want to import.

9. Click Save.

Uploading dictionaries from Watson Explorer oneWEX

You can import dictionaries from IBM Watson Explorer oneWEX.

1. From Watson Explorer oneWEX, Version 12.0.0 or later modifications or fix packs, download the dictionary CSV file.

   • Log in to the oneWEX administrator console.
   • Open the Resource tab.
   • Select dictionary enrichments, open the dictionary tab, and click the download icon. The dictionary is downloaded as a CSV file.

2. From your Discovery Content mining project, open the Content Mining application.

3. From the analysis view of your collection, click the Collections link in the breadcrumb to open the Create a collection for your analytics solutions page of the Content Mining application.

4. To create an annotator, click collection, and then select custom annotator from the list.

   

5. Click Create custom annotator.

6. Name your annotator, and then optionally add a description.

7. From the Annotator Type menu, select Dictionary, and then click Next.

8. Click the Import button, and then upload the CSV file of the dictionary that was downloaded from oneWEX.

9. Click Import, and then click Save.

Importing character patterns from Watson Explorer oneWEX

You can import character patterns from IBM Watson Explorer oneWEX.

1. From your Discovery Content mining project, open the Content Mining application.

2. From the analysis view of your collection, click the Collections link in the breadcrumb to open the Create a collection for your analytics solutions page of the Content Mining application.

3. To create an annotator, click collection, and then select custom annotator from the list.

   

4. Click Create custom annotator.

5. Name your annotator, and then optionally add a description.

6. From the Annotator Type menu, select Regular expression, and then click Next.

7. Click the Import button.

8. Select the JSON file that you want to import, and then click Save.

Importing facets from Watson Explorer Content Analytics Studio IBM Cloud Pak for Data

You can import a PEAR file to use as the machine learning source file from IBM Cloud Pak for Data deployments only.

You can show Content Analytics Studio facets in the Content Mining application. Only facets with a UIMA Feature of type Literal Value are displayed.

For more information about how to import Content Analytics Studio machine learning models for use in other project types, see Use imported ML models to find custom terms.

1. From the Watson Explorer Content Analytics Studio, export the machine learning model that defines the facets that you want to use. The model file must have a .pear extension.

2. In the export configuration, remove the facet path, but keep the subfacet value. Set the Index Field name to the Facet Tree Path in Content Analytics Studio.

   For more information, see Creating Custom PEAR Files for Use with Lexical Analysis Streams.

3. From your Discovery Content mining project, open the Content Mining application.

4. From the analysis view of your collection, click the Collections link in the breadcrumb to open the Create a collection for your analytics solutions page of the Content Mining application.

5. To create an annotator, click collection, and then select custom annotator from the list.

   

6. Click Create custom annotator.

7. Name your annotator, and then optionally add a description.

8. From the Annotator Type menu, select PEAR File, and then click Next.

9. Click Select file to find the .pear file that you exported.

10. Specify a facet path, and then click Save.