--- sourceDocument: Brazil IT Service Management sourceDocumentLink: https://www.servicenow.com/docs/r/it-service-management Release : - brazil ft:locale : - en-US ft:publication_title : - Brazil IT Service Management ft:clusterId : - itsm bundleId : - itsm workflow : - Technology --- # Manage # Managing Task Intelligence for ITSM models {#ariaid-title1} Release version: Brazil Updated September 10, 2026 ![](https://www.servicenow.com/docs/portal-asset/ico-clock) 2 minutes to read Summarize ![AI sparkle icon](https://servicenow.com/docs/portal-asset/ai-sparkle-icon) Summarized using AI This content was generated using new OpenAI-powered functionality. Results are provided on an as is basis and are not guaranteed to be accurate or complete. ## Summary of Managing Task Intelligence for ITSM models Task Intelligence for ITSM leverages machine learning to provide predictive recommendations and identify similar records for IT service incidents. These actionable insights appear in a side panel within incident forms, enhancing incident resolution efficiency and accuracy. Show full answer Show less ## Key Features * **Prediction Model Types:** * **Incident Categorization:** Predicts fields for new IT service incidents. * **Similar Incidents:** Finds similar incident records by comparing prediction and training tables. * **Major Incident Recommendation:** Suggests linking current incidents to active major incidents and proposes potential major incidents. * **Similar Open Change Requests for Incident:** Identifies related open change requests by comparing incident and change request tables. * **Similar Open Problems for Incident:** Finds related open problems by comparing incident and problem tables. * **Model Lifecycle:** Creating, training, assessing, deploying, editing, and exporting prediction models to optimize incident management workflows. * **Model Assessment and Testing:** Preview model performance on historical data during setup, use "Monitor only" mode for live data validation without modifying records, and track prediction accuracy over time with analytics dashboards. * **Exporting Models:** Enables reuse of trained models across different ServiceNow instances, saving time and effort. ## Practical Application for ServiceNow Customers By implementing Task Intelligence prediction models, customers can automate incident field predictions, identify relevant related incidents, changes, and problems, and improve major incident handling. This reduces manual effort, increases incident resolution speed, and enhances data consistency. The ability to test models before full deployment ensures confidence in predictions, while exporting models supports scalability across environments. Use the machine learning capabilities of Task Intelligence for ITSM to predict field level recommendations and similar records on the incidents which appear as actionable recommendations in the side panel. Managing Task Intelligence for ITSM involves the following tasks. ## Creating a prediction model {#using-task-intelligence-itsm__section_y33_djt_2yb} You can create and deploy solution-based prediction models to predict incidents fields and actionable real-time recommendations based on the similarities between two types of tables by comparing their fields for IT service incidents. Task Intelligence for ITSM provides the following types of prediction models: * Incident Categorization: This model predicts incidents fields for new IT service incidents. * Similar Incidents: The model looks at the prediction fields of a prediction table and the training fields of a training table. It uses the similarities in these fields to predict similar records for incidents. * Major Incident Recommendation: The model recommends similar active major incidents which the current incident can be linked to, and recommends you propose similar incidents as a major incident. * Similar open Change Requests for Incident: The model looks at the incident fields of an incident table and the change request fields of a change request table. It uses the similarities in these fields to predict similar change requests for incidents. * Similar open Problems for Incident: The model looks at the incident fields of an incident table and the problem fields of a problem table. It uses the similarities in these fields to predict similar problems for incidents. {#using-task-intelligence-itsm__ul_lbs_41s_zbc} Creating a model involves the following steps: * Set up a model: Set up a prediction model using a template available in the base system. * Define a model: Specify the purpose of the model by selecting the training table and prediction table which will be used for training the model further. * Train a model: Train a model to make predictions using your data. * Assess your model: Assess the results from the model training, view sample results for the predictions, and select the prediction preferences and behavior for your model. * Deploy your model: Deploy your model to predict incident fields on incident forms. {#using-task-intelligence-itsm__ul_mdg_w35_2yb} ## Testing a prediction model {#using-task-intelligence-itsm__section_testing_model} Before you deploy a model, preview how it performs on your data: * During setup, use the assess step to view sample predictions on past records. For more information, see [Assess the similarity model](https://www.servicenow.com/docs/9lzvPeGllpMTIZ6Z~VrYMA "Assess the results from the model training and view sample results to see the similar records predicted for incidents. Reviewing the results gives you a preview of how your model will perform after being deployed. Based on the sample results, select the prediction preference."). * To validate a model against live data without changing incident records, set the prediction behavior to Monitor only. The model runs in the background and stores predictions for you to review. {#using-task-intelligence-itsm__ul_testing_model} After deployment, track prediction accuracy over time in the analytics dashboard. For more information, see [Monitoring and Analytics for Task Intelligence for ITSM](https://www.servicenow.com/docs/KYTr417TLnnw5xYfEUqRRQ "You can view the impacts of your trained incident prediction models. Monitor model performance overtime, track business value, and view what predictions your agents did and didn't use."). ## Editing a prediction model {#using-task-intelligence-itsm__section_nzs_gjt_2yb} Edit a prediction model that has already been trained and deployed. Change the model configurations, view the updated training results, and redeploy the model. ## Exporting a prediction model {#using-task-intelligence-itsm__section_cmb_3jt_2yb} Export a prediction model in Task Intelligence for ITSM to another instance so you can use the model in the other instance without recreating the model from scratch. * **[Create an incident prediction model in Task Intelligence for ITSM](https://www.servicenow.com/docs/j3abtHQRNuXa0olZN6y6WQ)** Create and deploy solution-based incident categorization prediction models to predict incidents fields for new IT service incidents. * **[Create a similar records prediction model in Task Intelligence for ITSM](https://www.servicenow.com/docs/EtxTTcb9n4Jd5i4hJaOFQw)** Set up a training model to help it recognize similarities between two types of tables by comparing their fields. * **[Edit an incident prediction model in Task Intelligence for ITSM](https://www.servicenow.com/docs/N3R5oCjemloNTrTaxPrtgw)** Edit either the Similar Incidents or Incident Categorization model that has already been trained and deployed. Change the model configurations, view the updated training results, and redeploy the model. * **[Export an incident prediction model](https://www.servicenow.com/docs/yGXXgNmSAE2o8QzB5dIBdw)** Export either the Similar Incidents or Incident Categorization model model in Task Intelligence for ITSM to another instance so you can use the model in the other instance without recreating the model from scratch.