Based on its analysis of the content of a document in English, the Rosette categorization endpoint recognizes the following contextual categories:
Categories |
|
Arts & Entertainment |
Travel |
Business |
Automotive |
Education |
Careers |
Food & Drink |
Family & Parenting |
Hobbies & Interests |
Health & Fitness |
Law, Gov’t & Politics |
Home & Garden |
Personal Finance |
Pets |
Real Estate |
Religion & Spirituality |
Science |
Sports |
Society |
Technology & Computing |
Style & Fashion |
|
These are the contextual categories defined by the IAB Quality Assurance Guidelines (QAG) Taxonomy.
-
Before analyzing, Rosette filters out some stop words and punctuation, such as “the” “?” “a” “it”, to increase the accuracy of the analysis.
Rosette supports both singlelabel and multilabel categorization. By default, the categories endpoint is set to multilabel and will return all relevant category labels with a raw score above an internal threshold. In addition to a raw score, which can be any number from negative infinity to infinity, each category label is returned with a confidence score, which can be any number between 0 and 1.
To return only a single category label per document, set {"options": {"singleLabel": true}}
. Both a raw score and a confidence score will be returned. To override the internal threshold, -0.25, set scoreThreshold
to a value of your choosing.
More Info:
https://developer.rosette.com/features-and-functions#categorization