The similarity rule is used to measure the semantic similarity between a content and a reference text or set of texts.

​
Parameters:

  • type: similarity (specifies the rule type)
  • value: A string or array of strings containing the reference text(s) to compare against
  • expected: Usually set to pass to ensure the content is sufficiently similar to the reference
  • threshold: A number between 0 and 1 representing the minimum similarity score required (e.g., 0.8 for 80% similarity)