Lab Exercise 2 – Identify Optimal Values of Hyperparameters for iUnique

In this lab exercise, the main objective is to identify the optimal values of hyperparameters for iUnique (Unique Clustering) analysis. The procedure involves uploading data, starting the analysis, verifying the results, publishing the optimized hyperparameter template, and mapping it to the organization. This exercise aims to improve the recommendation of runbooks and ticket clustering for automation candidates.

Scenario

An organization has requested for further optimization of models for recommendation of runbooks and ticket clustering used for identification of automation candidates. The scenario will remain the same for all the exercises covered in this module.

In this lab, we will showcase the detailed procedure for analyzing and identifying optimal values of hyper parameters for iUnique (Unique Clustering).

Prerequisites

User should have Organizational Admin credentials.

Solution

  1. Open BigFix Runbook AI Web URL and login with Super Admin credentials.
  1. Go to Advance ConfigirationWorkbench and click Unique Analysis.
  2. Click Add New Analysis. It will bring up Upload Workbench Data page.
Figure 1. Figure 272 - Upload Workbench Data
  1. Select Organization and click Add Mock Organization. It prompts user to enter the Organization Name.
Figure 2. Figure 273 – Mock Organization
  1. Click Save.
  2. Select Module as ‘Incident’.
  3. Type the Analysis Name.
  4. Click Choose File to upload the sample tickets as per the provided template. The template can be downloaded by clicking Download Template.
  5. Select Hyper Parameter Template Version from the dropdown.
  6. Click Start Analysis to begin the analysis for Unique Clustering.
Figure 3. Figure 274 – Analysis Created Successfully

On clicking Start Analysis, one job is created for Unique Analysis and another for Recommendation Analysis. Newly added analysis is listed on the Manage Jobs page.

  1. To enable a job to view analysis, go to Actions  Manage Jobs.
  1. Select the newly added job for unique analysis with status as Queued.
  2. Click Enable Jobs.
  1. A confirmation message appears.
    Figure 4. Figure 275 - Confirmation Message
  2. This adds the analysis and lists it in Unique Analysis page with status as In Progress.
  1. Once the analysis is complete, status changes to Pending Verification. You can now validate the results.
Figure 5. Figure 276 – Unique Analysis
  1. To analyze the added unique analysis data, click next to analysis to be verified.
Figure 6. Figure 277 - Unique Analysis (cont.)

Once you have verified the analysis results, you can publish the most optimized hyperparameter template.

  1. Go to Advance ConfigurationWorkbenchUnique Analysis, click next to the analysis user want to publish.
Figure 7. Figure 278 - Unique Analysis (cont.)

User can publish only successful iterations.

  1. A confirmation dialog box appears. Click OK.
Figure 8. Figure 279 – Confirmation Message
  1. Publish the selected hyperparameter configuration template and map the published hyperparameter configuration template to organization under analysis. Click to map it to organization.
Figure 9. Figure 280 - Unique Analysis (cont.)
  1. Select the Template Name, Organization and Module from respective dropdowns.
  2. Click Save.
Figure 10. Figure 281 – Map Template with Organization
  1. After successful mapping of hyperparameter configuration template with organization.

Conclusion

Post the completion of this exercise, you should have a good understanding of identifying the most optimal values of hyperparameters for Unique Clustering Analysis following multiple iterations.

The next step is to perform the analysis to arrive at the optimal values of hyperparameters for iRecommend which will be covered in the next exercise.