Questions about Website Categorization Use | Website Categorization API | WhoisXML API


FAQs

How parked domains are categorized?

All the parked domains are assigned the “Internet and Telecom” category.

IAB Content Taxonomy

The Website Categorization API performs classification based on the IAB Content Taxonomy 2.2.

Initially, IAB taxonomy used up to 4 tiers for some categories. However, the lower-level categories are too narrow, so we combined the lower-level tiers and shrunk the taxonomy to 2 levels for better accuracy and readability. All the original IDs are kept without changes, so you can restore the full path from IAB files if necessary.

  • Tier 1: reflects the top-level category of the content. Usually, such categories are too broad and point only to a general content topic.
  • Tier 2: reflects IAB's tiers 2, 3, and 4. Such categories are narrower and describe content more specifically.

Usually, the Website Categorization API returns multiple categories sorted by the relevance in descending order. Without the "minConfidence" parameter, the API returns all the categories having a relevance greater than 0.5. All the results are sorted by their relevance in descending order. The relevance is calculated as the maximum of both tiers' probabilities. For example, if Tier 1's probability is 0.90 and Tier 2's is 0.99, the overall relevance of the set will be 0.99. And vice versa, if Tier 1 is 0.98 and Tier 2's is 0.8, the final probability is 0.98.

The complete list of the categories and the implementation guides is available at: