Big Text Workbench IBM Research

Big Text Workbench was an experiment in applying named entity recognition to book-length texts. People, organizations, and locations mentioned in the text would be automatically extracted, disambiguated, and counted, allowing visualizations such as a zoomable world map showing how many times places in that country are mentioned in the text.

Collaborators

Joan Morris DiMicco

Technologies

Java, Piccolo2D, and OpenNLP