CroNER Demo

CroNER is a tool for recognizing and classifying named entities in natural language texts in Croatian. CroNER recognizes the following classes of named entities: Persons (and Personal Possessives), Locations, Organizations, Ethnics, Date and Time, Money, and Percentage. Although developed for Croatian, CroNER can also be applied (with a small drop in performance) to texts in closely related languages such as Slovenian, Serbian, Bosnian, Montenegrian, and Macedonian.

CroNER was developed using a sequence labelling model, namely Conditional Random Fields (CRF). The model was coupled with well-engineered Named Entity Recognition features. Some of the features specifically account for morphological complexity present in most Slavic languages.

The model upon which the system was built is thoroughly described in this publication. If you use CroNER in your research, please reference this publication.

Enter some text (containing named entity mentions, e.g., news) in the box below and let CroNER recognize and classify named entities in the text. The color indicates the type of named entity and you can turn on/off annotation of certain named entity types.
Person Possessive Location Organization Ethnic Time Date Money Percentage