Lab Exercise 3 – Identify Optimal Values of Hyperparameters for iRecommend

This lab exercise covers the procedure for analyzing and identifying optimal values of hyperparameters for iRecommend (Recommendation). Users will learn how to enable the recommendation job, run new iterations for recommendation analysis, view and validate recommendation results, and publish the hyperparameter configuration template.

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 iRecommend (Recommendation).

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 Configuration Workbench and click Recommend Analysis.
Figure 1. Figure 282 – Workbench Recommendation Analysis
  1. Go to Actions  Manage Jobs and enable the Recommendation Job. It starts with prefix FetchUniqueRecommendation and includes your organization name.
Figure 2. Figure 283 - Recommendation Analysis (cont.)

Before enabling recommendation job, ensure that the unique analysis for the organization for whom you are adding the new iteration, is published. Then enable the recommendation job from the Manage Jobs page.

  1. Go to Workbench and click Recommendation Analysis.
  2. Click Run new Iteration to run a new iteration for Recommendation Analysis for Organization.
Figure 3. Figure 284 - Recommendation Analysis (cont.)
  1. A popup window for Template Version(s) appears.
Figure 4. Figure 285 – Template Version
  1. Select a template form the Hyperparameter Template dropdown list.
Figure 5. Figure 286 – Template Selection
  1. Click Run. A confirmation dialog box appears.
Figure 6. Figure 287 – Confirmation Message

The new iteration is added and appears at the bottom of the list in the grid below.

Figure 7. Figure 288 – Addition of New Iteration
  1. To view the recommendation analysis results, click next to analysis for Organization.
Figure 8. Figure 289 – View Analysis
  1. User can even validate and enrich recommendation results. Below figure shows list of relevant runbooks for corresponding ticket categories. If you are fine with the recommendation results, then proceed with publishing of hyperparameter configuration template.
Figure 9. Figure 290 – Enrich Recommendation
  1. Go to Advance ConfigurationWorkbenchRecommendation Analysis. Click next to the analysis for Organization to publish.
Figure 10. Figure 291 - Publish Analysis
  1. A confirmation dialog box appears.
  1. Click OK.
Figure 11. Figure 292 – Confirmation Message
  1. To publish the selected hyperparameter configuration template and map it with organization, click on icon.
Figure 12. Figure 293 – Import Parameter Template
  1. Select the Template Name, Organization and Module from the respective dropdown lists for mapping.
  2. Click Save.
Figure 13. Figure 294 – Map Template with Organization

Conclusion

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

This concludes this module covering the identification of optimal values of hyperparameters for Recommendation and Unique Clustering.

Let’s see how a user can configure and use the Document Process and Analysis functionality in the next module.

Related Documentation

BigFix Runbook AI Configuration Guide

BigFix Runbook AI Troubleshooting Guide