AI Similar Requests and Ticket Clustering
AI-powered detection of similar requests to support problem identification and pattern recognition.
Sera AI can identify requests that are similar to each other, supporting problem management by helping specialists spot patterns and recurring issues across incoming tickets. This capability works for any ticket category, not just incidents.
How similar request detection works
Sera AI looks across the Xurrent ecosystem for other requests that the specialist has permissions to access. If it identifies similar requests, it can suggest grouping them together and propose resolution steps based on past interactions with similar issues.
The suggestions are action-oriented and dynamic. The set of suggested actions is specific to the context of the current request. If grouping is identified as a reasonable choice, it is included in the suggestions. The AI only suggests next steps it considers truly actionable.
The ai_similar_request expression
The ai_similar_request expression is available for use in automation rules. It allows rules to evaluate whether an incoming request is similar to existing requests based on AI analysis of the request content.
This expression can be used to flag incoming requests that match a known pattern, route requests that resemble a specific category of issues to a designated team, or trigger notifications when a cluster of similar requests suggests an emerging problem.
Configuring automation rules for similar requests
To use AI similarity in an automation rule:
- Navigate to the Automation Rules section of the Settings console.
- Create a new automation rule or edit an existing one.
- In the conditions section, use the ai_similar_request expression to define similarity criteria.
- Configure the actions to take when a similar request is detected (for example, relate the request to a problem record, notify a problem manager, or apply a specific category).
Supporting problem identification
When multiple users report the same or similar issues in a short period, it often indicates an underlying problem. By configuring automation rules that use the ai_similar_request expression, organizations can surface these clusters early and initiate problem management before the issue grows.