Predictive Intelligence frameworks

  • Release version: Zurich
  • Updated July 31, 2025
  • 2 minutes to read
    • Summarize
    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 Predictive Intelligence frameworks

    The Zurich release of ServiceNow introduces three distinct Predictive Intelligence frameworks—classification, similarity, and clustering—each designed to enhance different predictive use cases. These frameworks leverage machine learning to improve automation and decision-making within your records and workflows.

    Show full answer Show less

    Key Features

    • Classification Framework: Automates the assignment of categorical field values based on historical data, such as categorizing incidents from their short descriptions. This reduces task resolution times, error rates, and the number of interactions needed to complete tasks by enabling automatic categorization and routing.
    • Similarity Framework: Identifies records with similar attributes to a new record, recommending resolutions by referencing similar past incidents. Its context-aware algorithms recognize synonyms and industry-specific terminology, helping agents quickly find effective solutions without requiring exact keyword matches.
    • Clustering Framework: Groups data into clusters to reveal patterns, detect major issues, or identify data gaps. For example, clustering can highlight a major outage by grouping similar incident records together.

    Key Outcomes

    • Enhanced automation of categorization and assignment processes, leading to faster and more accurate task resolution.
    • Improved service quality through intelligent resolution recommendations based on similar historical records.
    • Greater insight into data patterns and anomalies, enabling proactive management of incidents and service requests.

    Additional Notes

    The earlier regression framework for numeric predictions was deprecated in the Washington DC release; while existing regression models can still be trained and edited, creating new ones is no longer supported.

    Predictive Intelligence provides three different model frameworks in the Zurich release: classification, similarity, and clustering. Each framework specializes in different types of predictions.

    Predictive Intelligence classification framework

    The Predictive Intelligence classification framework enables you to use machine-learning algorithms to set categorical field values during record creation. For example, you can use the model to set the incident category based on the short description. You can train predictive models so they act as an agent to categorize and route work automatically based on your past record-handling experience.

    Enable Predictive Intelligence to handle volumes of incoming requests at lower costs. Automate the categorization and assignment of requests to reduce:
    • Task resolution times.
    • The number of interactions required to resolve tasks.
    • The error rates of categorizing and assigning work.

    For more information, see Create and train a classification solution.

    Predictive Intelligence similarity framework

    The Predictive Intelligence similarity framework identifies existing records that have similar values to a new record. For example, you can train a subset of your incident records to recommend a resolution based on the information of a similar incident record. By borrowing from similar closed incidents that have a proven resolution, you can help agents and fulfillers quickly provide the best resolution for an incoming incident.

    The similarity framework doesn't need an exact match of keywords for its text comparisons because its algorithms identify similar words and synonyms based on similar contexts. For example, the phrases printer not working and printer broken are both recognized as similar. The framework also collects, learns, and applies your industry-specific context. For example, the phrase unable to join network has a different context in a computer networking company than it does in a healthcare insurance company.

    The similarity framework uses a workflow similarity solution. For more information, see Create and train a similarity solution.

    Predictive Intelligence clustering framework

    Clustering divides data into groups that can then be used to identify patterns. You can then address records collectively or find gaps in existing data. For example, you can group similar new incidents to identify a major outage.

    The clustering framework uses a workflow clustering solution. For more information, see Create and train a clustering solution.

    Deprecated in the Washington DC release: Predictive Intelligence regression framework

    Important:
    Support for creating new regression solutions was removed in the Washington DC release. You can train and edit existing solutions, but you can't create new ones. This information is provided for legacy context.
    Regression is a machine-learning framework that uses historic data to predict numeric outputs, such as a temperature or a stock price.

    For more information, see Create and train a regression solution.