Setting up training for similar tickets

Set up training for this AI algorithm to enable the discovery of historical tickets to aid in the remediation of current tickets.

About this task

When an incident occurs, it can be helpful to review details for similar tickets from the past to help determine a resolution. This model aggregates information from past similar tickets. It can also extract the steps that are used to fix previous tickets, if documented.

Training setup of the algorithm is done by completing some setup tasks. When the training setup is complete, you need to go to a different part of the AI Model Management to train the model.

Before you begin

To train similar tickets, you must provide ticket information. There is no set minimum of how much data to provide; the AI trains on whatever amount is available.

However, the more ticket data you can provide, the better. Any data that is collected for historical training is available to use as data. You can also collect data from the live integration and as future tickets are created, they are processed automatically.

Qualifying tickets

Important: Starting with IBM Cloud Pak® for AIOps version 4.1.0, similar incidents, such as from a ServiceNow integration, are called similar tickets.

To pass data precheck, a minimum of five qualifying tickets are needed. Qualified tickets are closed:

  • The state and closed_at fields from the raw ticket schema indicate closure.
  • The close_notes field from the raw ticket schema includes meaningful resolution notes with actions and entities that the similar tickets algorithm can extract. If your tickets do not include resolutions, or if they include default entries like Closed by Caller, these values do not identify paths to remediation, and the similar tickets algorithm is unable to use the tickets to provide recommended actions through similar tickets.

The following required fields are needed for a ServiceNow ticket to be qualified for the similar tickets AI algorithm.

Required attributes for the similar tickets algorithm
Attribute Description
short_description Short description or title of the ticket.
number Identification number of the ticket.
opened_at Coordinated Universal Time at which the ticket started formatted as yyyy-mm-ddTHH:MM:SSZ.
state Current state of the ticket.
closed_at Coordinated Universal Time at which the ticket was closed formatted as yyyy-mm-ddTHH:MM:SSZ.
close_notes Comments or closed notes that talk about the list of actions that are taken, or comments that are added by the user who is attending to the ticket.

Running the task

There are multiple parts to this task:

Starting the training setup

Go to the AI Model Management and complete some setup tasks.

  1. In the Cloud Pak for AIOps home page, click the navigation icon at the upper-left corner of the screen to go to the main navigation menu.
  2. In the main navigation menu, click Operate > AI model management to open the AI Model Management.
  3. On the Training tab, click Set up Training within the Similar tickets tile under the Trainable AI algorithms section.

Checking data integrations

Review the information on this page. It explains what this AI algorithm is, how it will help in your production environment, and provides a list of integrations needed to generate a model.

  1. On the Integrations, check that at least one integration is listed.

    Note: If no integrations are listed, or you expected to see more integrations, click Go to Integrations to modify your integrations. For more information, see ServiceNow integration.

  2. Click Next to move to the next panel.

Schedule the training

Decide whether to run the training setup on demand or to schedule it to run on an ongoing basis.

  1. Proceed as follows:
    • To run the training setup on demand, ensure that Schedule to run is set to Off, and go to step 3.
    • To run the training setup on a schedule, set Schedule to run to Yes, and go to the next step.
  2. Schedule the run. Click this option to specify a schedule. You can specify a start date with an optional end date, a frequency, and a time based on Coordinated Universal Time (UTC).
    • Decide whether to run the training setup on demand or to schedule it to run on an ongoing basis.
    • Training loads existing qualified ServiceNow tickets into the similar tickets AI algorithm.
    • If you set the ServiceNow integration to continuous data collection, scheduling frequent trainings helps ensure that the similar tickets AI algorithm includes the new tickets in its recommendations.
  3. Click Next to move to the next panel.

Deciding how to deploy the training

The option to review training results before deployment is unavailable for this algorithm. After training is complete, the model is automatically deployed. To see similar tickets appear in your subsequent incidents, go to the Automations page and enable Default story query similar incidents service policy in the policy list.

Click Done to save the training setup for the algorithm.

What to do next

Now that the training setup is complete, you can train the algorithm. For more details, see Launching the training.