Visualization and Knowledge Discovery in Textual Data

French-Croatian bilateral project addressing efficient visualization of huge quantities of text data. The system allows visual tracking of named entities and topics in a defined time period.


There are large amounts of text data available digitally. Modern information extraction techniques are able to extract occurrences of named entities such as names of people and locations. Such information can be very valuable for discovering new facts (e.g., the relation between a person and an event). However, the large quantities of data are difficult to handle manually. One way to tackle this complexity is through text visualisation. The goal of this project was to develop advanced text visualisation methods that can be used to efficiently navigate through text data and discover new knowledge.

Project fact sheet

Participants: University of Rennes (Annie Morrin, Jean Hugues Chauchat); University of Zagreb – TakeLab FER (Artur Šilić, Bojana Dalbelo Bašić)
Duration: 2 years