SensAI Predict

What is SensAI Predict, and how does it work?

SensAI Predict gives you the foundation for an AI-powered workflow. This AI engine leverages machine learning to accurately predict the category of an incoming Issue and classify it accordingly. This provides your support team with a built-in classification system for Issues, and prevents them from having to manually classify common Issue types based on the […]

How do I prepare my data for the Predict Model?

SensAI Predict assigns Labels to Issues based on the data that you’ve provided to the Model in the form of a dataset. The performance of the Model and the accuracy of your Labels depends on the following factors: The quality of the dataset you provide (containing the user’s first messages and Labels) The amount of […]

How do I add a new Label to an existing Model?

To add new Labels, you will need to prepare a CSV file with user messages that correspond to the Label. Make sure there are at least 500 user messages for each new Label. You can find full instructions on how to prepare this CSV file here: How do I prepare my data for the Predict […]

How do I add Predict Labels into Automations?

You can use Predict Labels with Automations to assign Issues to Queues and Agent Groups, filter Issues for Smart Views, send automatic replies based on the Label, and more. If you have existing Automations that use Custom Issue Fields or tags to organize Issues into Queues and Smart Views, and would like to also use […]

How do I use Smart Views to filter Issues by Predict label?

To create or update a Smart View to filter Issues by Predict Labels, navigate to the Issues page. On this page, click the ‘+’ button next to the ‘Shared Smart Views’ or ‘My Smart Views’ section header. In the ‘New Smart View’ pop-up, under the ‘View Issues matching the following conditions’ section, select ‘Current Predict […]

What is the Confidence Threshold, and what value should I set for it?

To define the Confidence Threshold, we must first define what Confidence is in Predict. Confidence is a numerical value that is assigned to each Label while Predict is evaluating an Issue. The Confidence value is determined by Predict based on the likelihood that a given Label may be associated with an Issue based on the […]

How are Accuracy and Precision calculated in Predict?

Accuracy is calculated as the percentage of times that an Agent did not mark the predicted Label as wrong. Accuracy reflects the Model’s performance. Your Accuracy calculation includes all Labels within the Model. The formula for Accuracy is: (N-r) / (N) N: Number of Issues for which any Label was predicted r: Number of Issues […]

Where can I see Predict label data in Power BI?

You can review Predict label data in Power BI by creating a custom report using the Support Analytics content pack. When creating that custom report, you’ll be able to search and add the fields ‘Predict Label’ and ‘Predict Label ID’ from the Visualizations/Fields area. For full steps on creating a custom report in Power BI, […]

SensAI Predict Glossary

The following glossary includes all metrics as they are defined in SensAI Predict. To learn more about Predict, see What is SensAI Predict, and how does it work? What is it? What does it do? How does it work? Learn more SensAI Predict Our AI engine that leverages machine learning to accurately predict the category […]

How do I create and configure Models with Predict?

Predict allows you to automatically triage incoming Issues and assign relevant Labels to them. You can use the assigned Label to setup corresponding Automations, Smart Views, Advanced Search, and more. You can create models: At a domain level, where all Issues from all your support channels are processed and labeled using a single model.  At […]

How do I create multiple Models for a Language?

You may want to set up different workflows for similar Issues from different channels. For example, you may want to set up a custom workflow for Password Reset Issues from In-App Messaging, and a different workflow for similar issues from Email.  By creating app-specific models, you will able be to deploy different models to process […]