---
sourceDocument: Yokohama Data and Automation
sourceDocumentLink: https://www.servicenow.com/docs/r/yokohama/integrate-applications

 Release :

    - yokohama

ft:locale :

    - en-US

ft:publication_title :

    - Yokohama Data and Automation

ft:clusterId :

    - crint

bundleId :

    - crint

workflow :

    - Creator


---

# Select option for Recommended Actions application

# Select option for Recommended Actions application {#ariaid-title1}

* Release version: Yokohama
* 
* Updated January 30, 2025
* 
* ![](https://www.servicenow.com/docs/portal-asset/ico-clock) 1 minute to read

There are two options for the Recommended Actions application:
String-based and ML-based. These options decide the kind of data that you get for your search.
It displays data about automations in all states except the Published
state.

## Before you begin

Install the Recommended Actions application. For more information, see
[Install Recommended Actions application](https://www.servicenow.com/docs/_0DuH_C5Eut3Cb1vk~aOaQ "Install the Recommended Actions application to use its capabilities.").

Recommended Actions searches for matching string in Process name, Description, and Applications used fields of the Automation requests form,
and Automation name, Description, and Source in the Automation Attribute table.

* String-based: Searches for the exact string in the same order as in the search criteria.This is a good option if you want to use short phrases or single
  words as your search criteria. By default, this option is enabled.

* ML-based: This is helpful if you have too much data and your search criteria is long. To use the ML-based recommendations, you need at least 10 records in Automation Center.

{#select-option__ul_xm1_444_dwb}

Role required: sn_ac.automation_technical_user and
sn_nb_action.next_best_action_author

## Procedure

1. Navigate to AllRecommended ActionsRecommendations.
2. Mark the Active flag to true for the required options.  
   Note:  
   To identify the options, check the names. The names that have ML appended at the end are for ML-based recommendations.
3. If you have chosen ML-based option, then perform the following steps:
   1. Navigate to AllPredictive IntelligenceSimilaritySolutionDefinition.
   2. Open Automation request process name similarity and Automation attribute automation name separately.
   3. Select Update \& Retrain for each one of them.

   {#select-option__substeps_nmb_fq4_dwb}  
   After selecting Update \& Retrain, the training will start in the scheduler. After it's complete, it displays Solution complete.

*[\>]: and then