Named Entity Recognition, (NE), for Swedish, without the use of Gazeteers

NE is the process of identifying and classifying names of people, places etc. NE recognition is usually based on pre-stored lists of entities. The method exploited here is based on trigger words, typical predicates and constructions, associated with the entities, suffixes, prefixes and other simple heuristics. (For more information, look at the references at the end of this page.)

Select one or more of the entities to search for:

Money Expressions (e.g. 100 dollar) Organisations (e.g. UNESCO) Time Sequences (e.g. i tisdags)
Transportation Means (e.g. fartyget) Communication Means (e.g. internet)
Geographical Places (e.g. Stockholm)
Social Places (e.g. klubb) Narcotic Substances (e.g. droger) Persons (Pronouns included) (e.g. Kalle)

Select a file to analyze, (mostly newspaper articles):


Type in your own, tokenized example to analyze (max 2000 characters):
Text Field Use Menu

Kokkinakis D. (1998), AVENTINUS, GATE and Swedish Lingware, In Proceedings of the 11th NODALIDA Conference, Copenhagen, Denmark
The NE software is implemented by Dimitrios Kokkinakis
The WEB design by Sofie Johansson Kokkinakis
Last Updated: 19990314