Creating a Data collection for Knowledge Buddy
If you do not create a collection, all data sources that you create are assigned to Knowledge, which is the default collection for all data sources. This topic presents step-by-step instructions on how to create additional collections.
Creating a new collection
To begin creating a data collection, you must first log in to the Knowledge Buddy portal. You must have the Data Source Manage access role to create new data collections.
In the navigation pane, click Data collection. The Data collection landing page presents a list of all current data collections, together with the name of the operator who created and updated the collection, the date and time of the last update, and a short description of the collection.
In the upper- right corner, click Create data collection.
In the Create Data collection dialog box, in the Name field, enter a name for the new collection. A collection name cannot contain empty spaces, or any of the following symbols: forward slash (/) or back slash (\), question mark (?), the equal operator (=), or the ampersand (&) symbol.
In the Description field, enter a short description for the collection.
In the Access Configurations section you configure which roles have access to the collection. By default, only the Manage collection and View collection access types are added, but you can change them or click Add to add more as needed.
Nota: Data collections are secured by defining who can view and manage the collection. For more information about Knowledge Buddy security, consult the documentation at Knowledge Buddy security.
Next, select the Advanced settings checkbox to configure the content processing options for the collection. By default, all data sources added to the collection inherit the chunking strategy from the collection. You can also override the chunking strategy used by the collection by setting the strategy for each data source individually.
Select the Chunking method and adjust the values in the Chunk size and Chunk overlap fields, or use the default values.
Select the analyzers used for the different types of attribution, such as Content level attribution, Chunk level attribution, Auto filtering attribution as well as the Embedding attributes.
Click Submit.
You have now successfully created a new data collection.
Editing a data collection
On the Data collection landing page, you can click on the name of a data collection to view relevant details about it.
Click Edit to begin making changes to the data collection.
Edit the details of the collection as needed. Note that you are unable to change the name of a collection after the collection has been created.
After you complete the changes, click Submit.
You have now successfully edited the collection.
This Topic is available in the following Module:
If you are having problems with your training, please review the Pega Academy Support FAQs.
Quer nos ajudar a melhorar esse conteúdo?