---
sourceDocument: Australia IT Operations Management
sourceDocumentLink: https://servicenow-prod.fluidtopics.net/r/de-DE/it-operations-management

 Release :

    - australia

ft:locale :

    - de-DE

ft:publication_title :

    - Australia IT Operations Management

ft:clusterId :

    - itom

bundleId :

    - itom

workflow :

    - Technology


---

# Review similarity examples

# Review Event Management similarity examples {#ariaid-title1}

* Freigeben Version: Australia
* 
* Aktualisiert 12. März 2026
* 
* ![](https://www.servicenow.com/docs/portal-asset/ico-clock) 1 Minute Lesedauer

Review the similarity examples and scores that the system provides during solution training to see how the selected alert record compares to existing
alert records. For example, you can modify the similarity score threshold to increase the accuracy of your similarity recommendations.

## Vorbereitungen

Train a similarity solution in a ServiceNow datacenter. For more information, see [Create an Event Management similarity solution](https://servicenow-prod.fluidtopics.net/LHAj9XxDIRVfXLpdUIp7Xw "Create and train a solution that applies machine learning to a collection of words to target and suggest similar alerts in your instance dataset. For example, you can compare the text in a resolved alert record to an open alert record to reuse its resolution approach.").

Role required: evt_mgmt_admin

## Warum und wann dieser Vorgang ausgeführt wird

Adjust the current threshold to provide useful results. For example, a low similarity score has more coverage but less precision.

## Prozedur

1. Navigate to AllPredictive IntelligenceSimilaritySolutions.
2. In the ML Solutions list, locate your solution and click the Information icon (![Information icon]()).
3. Click Open Record.
4. Click the Similarity Examples related link.  
5. Review the similarity examples and their threshold scores to determine the accuracy and coverage levels you want applied to your solution similarity results.  
   The higher the similarity score, the more precise it is and the less coverage it offers. The lower the similarity score, the more coverage it has and the less precision it offers.
6. Based on your review, determine whether to increase or decrease the similarity score threshold value for your similarity solution  
   The solution returns only similarity results that have a score that is higher than the threshold. If the current threshold is not providing useful results, adjust the threshold value.
7. Navigate to Predictive IntelligenceSimilaritySolutions.
8. In the ML Solutions list, locate your solution and click the Reference icon (![Reference icon]()).
9. Click Open Record.
10. In the Solution Statistics tab, enter a new value in the Similarity Score Threshold field.  
    For example, suppose that the current score is 80. Based on your similarity example review, you might want to update the field by entering a higher score of 90, which will return fewer results but ones that are more similar to the alert record under review. A lower similarity score has more coverage but less precision.
11. Right-click in the form header and click Save.  
    Your solution uses the new similarity threshold value. In the example, the solution returns similar results that have a score higher than 90.
{#view-similarity-examples__steps_omc_lgk_yfb}