Use Cases for KB Article Generation
Use Cases and Functional Flow for Knowledge Generation
The following Use Cases are examples of customer requests for Knowledge Generation, based on Actions your can perform from the Content Generation > Knowledge Generation window.
Use Case 1: The user wishes to upload the document to the Aisera platform for immediate fulfillment consumption:
Go to Action -> Configuration.
Add the Ticket Data Source (which is mandatory) and choose other filters to select the required tickets for Knowledge Generation.
Go to the Pre-configuration window, select the Auto Publish checkbox, and the sub-option Publish to Aisera, then click OK.
Click on Generate Knowledge in the top right corner.
Process:
Knowledge will be generated and undergo a similarity check. You can see the cluster representatives and similar documents found under the Similar Documents column. Documents that are not found to be duplicates will be automatically uploaded to the Aisera knowledge repository.
Actions you can take:
You can review documents that are similar and either publish them to their external source system for further review or publish them to your Aisera platform for immediate fulfillment consumption.
You can upload documents already published to your Aisera platform into your knowledge repository by clicking on the Publish to SOR option, available on both the Knowledge Details page and the Similar Documents page.
Note: The Auto Publish option in the configuration automatically uploads the document to the desired source system. You still have the option to manually publish to any desired source system via the Similar Documents page and the Document Details page. Click on the document under Knowledge Cluster Representative, and the application will navigate to a page where you can see the document generated. After you see the generated document, you can perform the action to publish)
The user can also delete the document if they believe it is a duplicate; this action will remove the document from the particular job ID and it will not be shown under the knowledge generation tab.
When the user publishes the document to Aisera or SOR, the corresponding option will be disabled as publishing to one particular system can only be done once.
Use Case 2: You want to upload a document to your knowledge management system for further review. Follow the steps above but this time choose Publish to SOR under Auto Publish.
Process:
As knowledge is generated, it goes through a similarity check. You can see the cluster representatives and similar documents found in the Similar Documents column. Documents that are not found to be similar will be automatically uploaded to the configured external management system.
Actions you can take:
You can review documents that are similar and either publish them to their external source system for further review or publish them to the Aisera platform for immediate fulfillment consumption.
You can upload documents already published to SOR into the Aisera knowledge repository by clicking on the Publish to Aisrea option, available on both the knowledge details page and the similar documents page. Note: The Auto Publish option in the configuration automatically uploads the document to the desired source system. You still have the option to manually publish to any desired source system via the Similar Documents page and the Document Details page. Click on a document under Knowledge Cluster Representative, and the Aisera platform will navigate to a page where the you can see the generated document. From there, you can choose the Publish action.
You can also delete the document if you think it is a duplicate; this action will remove the document from the particular job ID and it will not be shown under the knowledge generation tab.
When you publish the document to Aisera or SOR, the corresponding option will be disabled because publishing to a particular system can only be done once.
Use Case 3: You don't want the generated documents to be consumed by the Aisera platform for fulfillment or uploaded to the knowledge repository automatically. Follow the steps mentioned above but this time choose the option, Manual Publish.
Process:
As Knowledge is generated, it goes through a similarity check. You can see the cluster representatives and similar documents found in the Similar Documents column.
None of the documents will be uploaded to Aisera or an external knowledge repository at this stage. You can review the generated documents on the Knowledge Generation page and decide whether to publish them to any of the available systems.
Actions you can take:
You can review all the generated documents and decide whether to publish the document to Aisera or SOR using the buttons in the Similar Documents page or the Knowledge Details page. Click on the document under Knowledge Cluster Representative, and the Aisera platform will navigate to a page where the you can see the generated document. You can then choose the Publish action.
Additional Use Cases
0 Cluster Scenario:
There is a possibility that there will not be any similar tickets found resulting in zero clusters. This situation is more likely to occur when a very small number of tickets are inserted as input, where no clusters can be formed. This is because the absence of clusters indicates a lack of similar tickets.
Handling Job Failures and Next Steps:
The current process primarily involves generating Knowledge documents, followed by a similarity check, auto-approval, and auto-commit. However, if any job fails after the Knowledge Generation step, during the next run triggered by the user, the similarity check, along with auto-approval and commit, should be rerun for the previous unsuccessful runs in addition to the new job run.
Even if the job fails during the auto-commit or auto-approval state, it's crucial to rerun the similarity check. This is because we continuously ingest documents, and it's vital to use the latest knowledge repository to validate duplicates and avoid committing duplicate documents.
Adding a Status Field for Knowledge Written to a Source of Record (SOR)
You can add a Status column in the Generated Knowledge table to track the progress of documents that you are writing back to your Source Of Record (SOR).
To add a Status Column:
Navigate to Content Generation → Knowledge Generation.
Click on the three dots in the top-right corner of the table grid.
Select Add/Remove Columns and add the new column called Status.
Select the documents you want to publish to the source system (such as, Salesforce, ServiceNow, Confluence).
Go to Actions → Publish to SOR.
Click on the refresh icon in the Status column to view the progress.
Status Indicators:
Publishing Underway – Document publishing is in progress.
Published – Document has been successfully published.
Failed – Publishing was unsuccessful.
Scenarios Involving Incremental Ticket Ingestion and Solution Identification
There are two scenarios that you might come across where no clusters are generated:
When you click on the incremental option and no new tickets have been ingested, an acknowledgement is displayed upon job completion. The message below indicates that there are no new tickets, resulting in the absence of formed clusters
Additionally, there might be a scenario where you have ingested the tickets, but the Aisera platform hasn't identified a solution from any of those tickets. To address this, the same acknowledgment message is triggered to inform you that no solution has been identified.
Last updated