Select option for Recommended Actions application
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.
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.
Role required: sn_ac.automation_technical_user and sn_nb_action.next_best_action_author
Procedure
-
Navigate to All > Recommended Actions > Recommendations.
-
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.
-
If you have chosen ML-based option, then perform the following steps:
-
Navigate to All > Predictive Intelligence > Similarity > Solution > Definition.
-
Open Automation request process name similarity and Automation attribute automation name separately.
-
Select Update & Retrain for each one of them.
After selecting Update & Retrain, the training will start in the scheduler. After it’s complete, it displays Solution complete.
-
Parent Topic:Configuring the Recommended Actions application with Automation Center
---
title: Select option for Recommended Actions application
description: 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.
locale: en-US
canonical_url: https://www.servicenow.com/docs/r/australia/application-development/automation-center/select-option.html
release: australia
product: Automation Center
classification: automation-center
topic_type: task
last_updated: "2026-09-10"
reading_time_minutes: 1
breadcrumb: [Configuring Recommended Actions, Configure, Automation Center, Discover, AI Workflow Factory Prime, Agentic app development, Building applications]
---
# Select option for Recommended Actions application
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://raw.githubusercontent.com/ServiceNow/ServiceNowDocs/australia/markdown/australia/application-development/automation-center/install-ra.md).
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.
Role required: sn\_ac.automation\_technical\_user and sn\_nb\_action.next\_best\_action\_author
## Procedure
1. Navigate to **All** > **Recommended Actions** > **Recommendations**.
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 **All** > **Predictive Intelligence** > **Similarity** > **Solution** > **Definition**.
2. Open **Automation request process name similarity** and **Automation attribute automation name** separately.
3. Select **Update & Retrain** for each one of them.
After selecting **Update & Retrain**, the training will start in the scheduler. After it’s complete, it displays **Solution complete**.
**Parent Topic:**[Configuring the Recommended Actions application with Automation Center](https://raw.githubusercontent.com/ServiceNow/ServiceNowDocs/australia/markdown/australia/application-development/automation-center/config-ra.md)
This file has no frontmatter.