Step 2 - Pre-Generation Parameters
The second step for Knowledge Generation is the setting the Pre- Generation configuration for the resulting knowledge document that you want to create.
Select the Pre-Generation Configuration option.
Choose either Auto Publish or Manual Publish.
Each of these options is described in the next sections.
Auto Publish: This option automatically uploads documents without the need for manual review to either the 'Aisera Knowledge Repository' or an external knowledge management system such as Salesforce, or ServiceNow.
After selecting Auto Publish, two additional sub-options will be displayed. You can choose either one or both.
Publish to Aisera: This will automatically upload the document to the Aisera Knowledge Repository for fulfillment purposes.
Publish to SOR: This will automatically upload the document to the configured external repository in a draft state, allowing you to review the document in your knowledge repository, before proceeding through their review cycle, and then publishing it. The Aisera GenAI Platform can then consume and use the document as an option for Fulfillment.
You have the ability to customize the error messages that can occur when publishing a document to your SOR (Source of Record) system fails due to missing or incorrect configurations.
To publish a document to an external knowledge management system, you must create a Workflow with the appropriate endpoints and event triggers. When a bot user clicks the ‘Publish to SOR’ button, the Aisera Gen AI Platform checks to determine if the necessary prerequisites are in place.
There are two primary scenarios that trigger an error message:
You have not created any required flow/event (a prerequisite for publishing).
You have created the event/flow, but with incorrect or incomplete configurations.
The Aisera Gen AI Platform now displays targeted error messages for each of these scenarios, enabling you to identify and resolve configuration issues quickly.
Scenario 1: No Event/Flow Created
When you click the Publish to SOR button in the Knowledge Generation window without having the required workflow/event, the platform will now display acknowledgement popups with clear error messages.
These messages will also appear in the Document Status column for visibility.
Acknowledgement Messages for ‘Publish to SOR’:
No Event Triggers Found (Either Not Created or Inactive): Use case: No events have been created, or the existing ones are inactive. Error message:
No active event found. Please create and activate an event using the right template flow to publish the knowledge.Event Triggers Exist Without a 'Template' Triggering Condition: Use case: An active event exists, but the required template trigger condition has not been configured. Error message:
Template trigger condition is missing for the active event. Please configure it to proceed with publishing.Multiple Event Triggers Found for the Same Template Condition (Active): Use case: More than one active event is configured with the same template condition (such as, multiple workflows using the "KCS" template), which is not allowed. Error message:
Multiple active events found for the same template. Please ensure only one active event exists per template to proceed with publishing.
These enhancements ensure you receive real-time, actionable feedback when publishing prerequisites are not met.
Scenario 2: Event Exists, but Configurations Are Incorrect If the user has created the required event, but there are configuration issues, the system will display failure messages under the status column when the user attempts to publish:
Workflow Configuration Issues Error message:
Workflow configuration error: Please review and update the configuration before retrying.External/API Errors Error message:
The system will show the specific error message received from the external API, providing visibility into the failure reason.
To use Auto Publish:
Navigate to Content Generation > Knowledge Generation
Navigate to Content Generation > Knowledge Generation.
Go to Actions > Configuration under Conditions, add the Job Name, select the ticket conditions, and click on Pre-Generation Configuration.
Under Auto Publish, there is an option called Publish to SOR. Select this option if you wish to automatically upload the generated non-similar documents to the configured knowledge management system.
Click OK, then click the Generate Knowledge button.
Manual Publish: This option generates the document, but does not upload it to the Aisera GenAI Platform or an external knowledge management system. Use this option if a Admin must review the document and can then manually publish it to the Aisera platform or an external knowledge management system.
There is an option to select a template to generate the documents 'KCS' and 'Default,' which can be chosen from the 'Generated Knowledge Template' drop-down menu. 'Default' will be selected if no other option is specified.
Once the configurations are done, click on ‘OK’.
When the user clicks on ‘OK’ - an automatic Data source called ‘Generated Knowledge (BOT ID - <#BOT ID>)’ will be created and attached to the specific bot to store the generated knowledge.
Knowledge Generation Rules:
The Generated Knowledge Data Source cannot be detached from the bot, meaning that you don't have the ability to remove the Data Source from the application.
This Data Source cannot be re-used for other applications. You can not attach this Data Source to other applications.
You have the ability to delete the Data Source by navigating to the Settings >Data Source page and clicking on Delete. This will remove the corresponding Data Source link in the application.
If you delete the original Generated Knowledge Data Source, if you go to the Knowledge Generation page, try to add the configuration again and click OK, the same ‘Generated knowledge (BOT ID - <#BOT Id>)’ Data Source will be created again. The previously generated documents will not be available in this new Data Source.
Attachments can be Ingested as Knowledge Bases in Knowledge Learning
File attachments (such as PDFs) can be ingested as separate KBs during Knowledge Learning. You can activate the option to ingest these attachments as autonomous files using the processFiles parameter in the Data Source configuration.
Mapping Knowledge Fields to Generated Documents
You can now upload documents to a specified source system using a JSON file, and then map fields that you want to add to each document when you generate knowledge articles. Assigning values to fields is often used to ‘tag’ groups of tickets or knowledge articles into categories.
Prerequisites:
Identify the fields and their corresponding values that you want to work with on the Knowledge Field Assignment page.
NOTE: The External System Type is no longer mandatory.
This will allow you to pre-select the fields and their values while generating the document.
Collect the following data points:
Field Name
Field Path
Data Type
Values within the Field
Selection Type (Single-select or Multi-select)
Provide these details to your Aisera Customer Success Team, so they can create a JSON file and upload it to Aisera.
Load the JSON file
To upload the .json file:
1. Navigate to Settings -> Configuration -> Knowledge Generation.
2. Add the External System Type. This section will appear blank if no files have been uploaded previously.
It lists the integrated systems for which JSON files have been uploaded in the past.
3. Click + Add a JSON File and choose the file to upload. A popup will appear. Select the integration type and upload the JSON file.
To Adding New Fields or Values:
To add new fields or values to existing fields, first export the most recently uploaded JSON file from the External System Type.
Edit the exported JSON to include the new fields or values.
In the pop-up menu, select the option, Add Delta Values from the JSON and click OK.
To Override Existing Data:
To completely replace the previously loaded data, download the JSON for the specific integrated system.
Modify it with the required fields and values, and select the Override option.
This action will entirely replace the fields and values, and the updated data will be reflected in the Knowledge Generation -> Action -> Configuration -> Knowledge Field Mapping page.
NOTE:
Overriding data may impact past runs. For instance, if a job was executed last month using specific field mapping and values, overriding the data will invalidate those fields and values. As a result, when uploading documents to the source system, these fields and values will be ignored, which is expected behavior. It is important to consider this potential impact when selecting the override option.
There is no direct option to update existing fields or values. To make updates, use the Override option without modifying other fields. Add the updated field or values to the JSON file, select Override, and upload it. This will retain the previous fields and values while incorporating the updated data.
Limitations:
Currently, all fields and values added are displayed as drop-down options in the Knowledge Field Mapping screen.
This approach currently has the following limitations:
1. Date Fields:
You must manually add date values to the JSON file..
2. Date/Time Fields:
Similar to date fields, you must manually add date and time values.
3. Open Text Fields:
You cannot currently enter free text into open text fields loaded from the JSON on the Knowledge Field Mapping screen in the Knowledge Generation page. All that they need to do is to add the values as well for the free text.
For example, the field Product Line, which is of type Text and accepts free text, requires that you define the possible values in the JSON file (such as ‘Analytics and Reporting’ or ‘Finance’). These values are then displayed as drop-down options for the Product Line field.
You can select a value based on the chosen category, since the Product Line field is only available when a value is selected in the Category field. Please note that there is no hierarchical display, and the Product Line field is currently mandatory when you select a category.
4. Multi-Selection Fields:
Fields that allow multi-selection in the source system are not supported at this time.
We are actively working on addressing these limitations.
Field Mapping for KB Articles is not mandatory. If there is a requirement to pre-select the knowledge fields while uploading the document, first select the ‘External System Type’ (such as, ServiceNow, Salesforce, or Zendesk) to which the JSON mapping is configured in the Tenant configuration screen (as mentioned above).
Date/Time Picker for Knowledge Field Mapping
You can now configure your Knowledge Field Mapping in alignment with the field types and configurations set in your source system, such as Salesforce, ServiceNow, or Confluence.
For each data type:
Date Field: Shows a date picker.
Date/Time Field: Shows a date/time picker.
Free Text Field: Displays a free text input box.
Multi select : Allows you to choose multi select in the drop down values.
The Aisera Gen AI system supports seamless mapping and value addition for all other data types, maintaining consistency with the source system's configurations.
Last updated