Most of the information available today is in free form text. Current technologies (Google, Yahoo!) allow us to access text only via key-word search.

We would like to facilitate content-based access to information. This requires programs that can, at some level, understand natural language text – categorizing documents by topic or function, retrieving entities and concepts – rather than looking for specific strings in text.

Some of the demos below directly address such Information Extraction tasks (e.g. Wikifier, Named Entity Recognition, Dataless Classification), while others exhibit fundamental natural language technologies that can support higher-level applications (e.g. Part-of-Speech Tagging, Shallow Parsing, Semantic Role Labeling).

If you like what our demos can do, and you’re interested in using these technologies more intensively, please visit our Software Page to download and run our packages yourself.

For the complete CogcompNLP Pipeline Demo, see

The Coolest

Natural Language Analysis

Entities and Information Extraction


Natural Language Applications




All Demos by Popularity

Name Views

All Demos by Title