---
sourceDocument: Australia Customer Service Management
sourceDocumentLink: https://servicenow-prod.fluidtopics.net/r/pt-BR/customer-service-management

 Release :

    - australia

ft:locale :

    - pt-BR

ft:publication_title :

    - Australia Customer Service Management

ft:clusterId :

    - csm

bundleId :

    - csm

workflow :

    - Customer and Industry


---

# Task Intelligence Admin Console

# Task Intelligence Admin Console {#ariaid-title1}

* Versão de lançamento: Australia
* 
* Atualizado 12 de mar. de 2026
* 
* ![](https://www.servicenow.com/docs/portal-asset/ico-clock) 1 min. de leitura

Use the Task Intelligence Admin Console to create, train, and deploy machine learning models that predict different types of information for case and
interaction records.

From the Admin Console, you can set up predictive models, preview the agent's experience, view when the models are active, and track model performance.  
The Admin Console provides tools that you can use to create and implement machine learning models in just a few steps. Each model follows a six-step process.

|-|-|
| 1. Select a model to use as a starting point. | Select the model based on what you want the model to do. For example: * Use the field prediction model to make field value predictions that you can use to categorize cases and interactions. * Use the case sentiment model to predict sentiment at any point from case creation to case resolution. {#csm-task-intel-admin-center__ul_tsd_bnv_c5b} |
| 2. Define the purpose of the model. | Tell the model when you want it to make predictions and what you want it to predict. For example, predict the category and priority when a new case is created. |
| 3. Select the data used to train the model. | Train the model using selected data, such as the text in the case short description and description, so it can learn patterns in the data. Then test the model to see how well it works. |
| 4. Assess the model's results. | View test results to see how your well the model performed. These results indicate how a model will perform after being deployed. |
| 5. Select preferences for prediction results. | Add predictions directly to record fields, show predictions as recommendations, or monitor predictions in the background. |
| 6. Deploy the model. | Review your selections and start using the model. |
[Tabela 1. Steps to create and implement machine learning models]

{#csm-task-intel-admin-center__table_rt4_2hv_c5b}  
You can also use the Task Intelligence Admin Console to access related applications. For more information about using the console, see the following topics:

* [Create a model to predict record
  fields](https://servicenow-prod.fluidtopics.net/RQFxyMoUeve5~QMdIOHEgw "Create and train a model to predict fields for case and interaction records.")
* [Create a model to predict
  case sentiment](https://servicenow-prod.fluidtopics.net/aWxvgr2eRRL6LhmKttJgtw "Edit and test the pre-trained sentiment model to predict sentiment for customer service cases.")
* [Create a model to detect
  case language](https://servicenow-prod.fluidtopics.net/B9YdEGxjg1LuDl6uu3dm1Q "Edit and test the pre-trained model to detect the language used to create customer service cases.")
* [Create a Document Intelligence use
  case](https://servicenow-prod.fluidtopics.net/O1kcKBV092POJCYBG13M5A "Create a use case that identifies the information to extract from email and case attachments and determines how users with CSM agent roles interact with the extracted values in the Document Intelligence workspace.")
{#csm-task-intel-admin-center__ul_sqk_y1r_fvb}
* **[Machine learning model setup and behavior](https://servicenow-prod.fluidtopics.net/cEWABq0u~IemRrK9B1yqVA)**   
  Set up models to predict field values and sentiment for customer service cases.