Projektnummer

0059938

Zusammenfassung

This project combines the latest scientific findings in natural language processing and information visualization in a powerful exploration tool that provides accelerated access to text document collections for investigative data-driven journalism. Whereas today's data-driven journalism is mostly based on structured data, i.e. data available in spreadsheets or databases, this project applies the principles of data journalism to unstructured text documents. Using statistical methods of language processing, such as named entity recognition and keyphrase extraction, important elements fro...