Thank you for your feedback.
Form temporarily unavailable. Please try again or contact docfeedback@servicenow.com to submit your comments.

Configure the similarity analyzer with Agent Workspace

Log in to subscribe to topics and get notified when content changes.

Configure the similarity analyzer with Agent Workspace

Use the similarity analyzer to define patterns and trends that can help your users resolve a record.

Before you begin

Role required: admin

About this task

Use Predictive Intelligence similarity solutions for the similarity analyzer to reference trends. Trends include common resolutions, linked problems, linked changes, attached articles, major incident detection, and linked major incidents. If you enable Agent Assist recommendations, you can provide helpful recommendations to agents when a trend arises.

Procedure

  1. Navigate to Similarity Analyzer > Trend Definitions.
    You can also use the Agent Workspace Guided Setup Search > Similarity Analyzer > Configure.
  2. Click New to create a new trend definition or open an existing trend definition.
  3. On the form, fill in the fields.
    Table 1. Trend Definition form
    Field Description
    Name Name of the trend definition.
    Table Table where you want to identify a trend.
    Trend type Type of trend configuration:
    • Common value - Identifies common values across a group of incidents. For example, say that the same value is for a resolution code.
    • Common reference - Identifies common related records in a group of incidents. For example, say that there are multiple incidents with the same problem record.
    • Common condition - Identifies a condition that applies to a group of similar records. For example, say that multiple incidents apply to the same location and are all high impact.
    Common field If the trend type is common value, select that field to check for duplicate values across similar records.
    Reference type If the trend type is common reference, select the type of reference relationship. Choices include simple reference or M2M reference. A simple reference refers to 1-to-many or 1-to-1 relationships. For example, an incident relates to a problem, but a problem might have multiple incidents that are related to it. An M2M reference refers to a more complex relationship, such as the relationship between incidents to knowledge articles. An incident might reference many knowledge articles and a knowledge article might have many references from incidents.
    Trend field If you select a simple reference, select the field to check for trend reference across similar records.
    M2M table Table that you would like to identify trends in. For example, select the m2m_kb_task to look for knowledge articles that are applied to a task.
    M2M from field Field that is related to the current table. For example, if you are looking for common articles that are related to an incident, this field is task.
    M2M to field Field that is related to the reference table. For example, if you are looking for common articles that are related to an incident, this field is kb_knowledge.
    Condition If the trend type is common condition, use the condition builder to create a condition to apply to a group of similar records. For example, set conditions for multiple incidents in this group that apply to the same location, are high impact and are in the same category.
    Application Scope that this trend definition should run in.
    Trend ID ID for this trend.
    Order Integer value that determines the priority of this trend versus all other trends. A lower value means that there is a greater priority for the trend. If no value exists, the priority goes by alphabetic order of the trend names.
    Threshold Number of duplicate values to indicate a trend.
  4. Click Submit.
Feedback