Skip to content


Developing a text intelligence model requires an adequate number of carefully chosen documents representative of all cases that the model must be able to predict.
Corpora are a powerful aid in the selection of documents for your projects.

A corpus is a collection of documents that, at upload time, get analyzed with Natural Language Understanding (NLU) software and indexed on the features that text analysis discovers. Indexing allows you explore corpus documents to quickly determine whether they fit your project needs and possibly use the corpus as a source of documents for other projects' libraries.