The machine learning service allows users to build machine learning models to:

• classify incoming items (a.k.a classification)
• score items based on relevance (a.k.a regression)
• cluster items together (a.k.a. topic modeling)
• extract important people, organizations, places, and whole phrases (a.k.a. named entity recognition)

For an introduction for working with the machine learning service, we recommend following our series of walkthroughs.

The backbone of the machine learning service is our own specialized natural language processing library, libNLP, for which we provide complete API documentation.