Version 0.1, November 15, 2009.
- This service detects the Topic central to a page.
- Topics currently include finance, sport, health, science, entertainment, technology, world news, us news and others.
- More topics and entities to be deployed: games, media, ...
Interpreting the resulting JSON record:
For product classification see http://products.speedi.ly/.
- Topic: the main Topic in the document.
The score is an absolute measure as well as a weighting between topics.
- language: the main language in the document.
Language detection covers main European languages, CJK, Russian, Thai...
No entity is detected when the language is not English.
- nsfw: number of distinct / total offensive terms.
- gory: A measure of the violence level in the document.
- entropy: A measure of the quality of the document.
- entities: Currently empty. Will contain entities extracted from the document.