News Taxonomy

A comprehensive classification system for English-language news content

• AP News Taxonomy includes standardized subjects, planned and breaking news events, people, organizations, geographic locations and more, all designed with news content in mind. Frequent updates ensure timeliness, and a rich network of semantic relationships between concepts enables creative solutions for content linking, search and discovery.
• Mappings to the IAB Tech Lab Content Taxonomy can be added to your service to augment your contextual advertising.
• Subscribe to the full suite of our AP Vocabularies, or just to the subsets of the taxonomy that you need.

How can I use it?

Integrate the News Taxonomy into your publishing systems to support manual tagging, or use it in conjunction with the AP Tagging service to apply the taxonomy to your content.

In addition to standardized terminology and unique IDs, the taxonomy stores a variety of details about the people, places, and topics it contains.

This information can power enhanced search experiences, browsing and discovery, or informational displays; for example:

• Synonyms, acronyms, and spelling variants for greater recall and findability.
• Properties of people, places and things – such as an athlete's uniform number, the latitude and longitude of a geographic location, or the stock ticker symbol for a company.
• Relationships between concepts, such as between a person and a geography (Politician-State), or between two organizations (Sports Team-Sports League).
• Hierarchical structure for subjects and geographic locations, to enable both broad and narrow searches.
• Mappings to IAB Tech Lab Content Taxonomy for contextual advertising support.

What is included in the News Taxonomy?

AP News Taxonomy

The AP News Taxonomy has five main areas of coverage:

• AP Subject
  A set of topics geared toward news coverage, ranging from broad categories (such as Politics or Education) to specific concepts (such as Voting rights or School curricula). Also includes many named events such as Academy Awards and Tour de France, as well as breaking news events as they happen.
  
  
• AP Geography
  Geographic place names arranged hierarchically – continents, world regions, countries, territories, national capitals, major world cities, US states, Canadian provinces, and a large number of US cities and towns.
  
  
• AP Organization
  Organizations and institutions from a wide variety of sectors: government organizations, non-profits, sports teams, colleges and universities, political and ideological groups, cultural institutions, and more.
  
  
• AP Person
  Celebrities, artists, designers, authors, business leaders, political figures, sports figures, royalty, and other newsmakers known at the global or US national level. Coverage is especially broad for US newsmakers in politics, entertainment and sports, including complete rosters for major professional sports teams, NCAA football and basketball athletes and coaches on the rosters of Power Five Conference and the Big East teams, all US officeholders at the federal and gubernatorial levels, and all candidates for those offices.
  
  
• AP Company
  Publicly-traded companies with shares trading on the following stock exchanges: NYSE, NASDAQ, Toronto Stock Exchange, Tokyo Stock Exchange and XETRA - Frankfurt. Also included are newsworthy companies trading on other exchanges and select private companies.

IAB Tech Lab Content Taxonomy Mappings

The IAB Tech Lab Content Taxonomy is the Interactive Advertising Bureau's standard collection of content classification categories. The taxonomy service gives you access to mappings between the AP News Taxonomy's Subject terms and the IAB's classification categories.

What about taxonomy updates?

The AP News Taxonomy is constantly being updated to capture the latest news and the biggest newsmakers. Whether it’s this week’s IPOs or the new crop of college athletes, AP’s taxonomy developers are always working to keep the vocabularies current and relevant.

Subscribers are kept up-to-date in real time – as soon as a change is published, the new version becomes available in the AP News Taxonomy. A detailed log of all changes is accessible through a separate API. You can keep track of all changes, or just the ones you care most about.

Taxonomy mappings are updated as external taxonomy updates become available, and new mappings are completed.

How does it work?

Subscribers can access the taxonomy by making calls to an API and request the full set of terms in a given vocabulary, a subset of terms, or information about a particular term. Calls are also available for retrieving deprecated terms, term change logs, and additional information about the structure of the taxonomy. Taxonomy data can be returned in Semantic-Web compatible formats, such as RDF/XML, RDF/TTL or JSON-LD. Information about a particular term can also be returned as HTML.

To learn more, refer to the Developer's Guide.

Tagging Service

Enrich your content with tags from AP News Taxonomy or other standard taxonomies

AP Tagging service receives your English-language news content and automatically returns relevant metadata, using standardized terminology from the AP News Taxonomy or the IAB Tech Lab Content Taxonomy.

The service goes well beyond mere text extraction; it uses a hybrid approach to concept and entity tagging, leveraging a deep knowledge graph for better domain targeting, as well as human-created context-aware semantic rules that understand your content and identify the most pertinent entities and topics.

Each tag applied to your content has a relevance score indicating how pertinent each tag is to your submitted text. Only the tags that truly describe your content will be returned.

Why human-created rules?

Human-created rules allow for more precise control over the service performance.

The system will recognize and return specific entities that it finds in the submitted content and also uses human-created semantic rules to identify topics that may not be explicitly mentioned in the text at all.

For example, a story about a particular country music star can trigger the "Country music" subject, even if the word "music" does not appear in the story.

What types of metadata are returned?

The tagging service analyzes each piece of submitted content and returns standardized names, relevance scores, IDs and other properties for all the relevant taxonomy terms that triggered a semantic rule. You can specify which taxonomy you want applied - AP's or IAB's - and which of the five areas of AP News Taxonomy's coverage the service should return.

The metadata output is enhanced with additional data properties. For example, companies are given a ticker, athletes are associated with their teams, and geographic locations get latitude and longitude data. You can also access the AP News Taxonomy or the IAB Tech Lab Content Taxonomy mapping for additional information about any given tag.

Additional tagging features will be added on an ongoing basis - please check back here for updates.

How does it work?

Subscribers can access the tagging service by making calls to an API either synchronously, or by submitting multiple documents simultaneously and retrieving the tags at a later time. Content may be submitted as plain text or XML.

Tagging service data can be returned in a Semantic-Web compatible formats, such as RDF/XML, RDF/TTL, n-triples or JSON-LD. It can also be returned in Simple XML format.

To learn more, refer to the Developer's Guide.