What is automatic text extraction?

Automatic text extraction is closely related to automatic text summarization.
Automatic text extraction extracts from a text relevant information which a user
searches for.
There are various techniques to carry out it this task. The techniques are language dependent and we have constructed systems which deals with Swedish language.

One technique is to use a tagger which tags each word with word class and from this information extract information. Try our Granska tagger.

One other technique is to look for Named Entities: person names, company and organizations names, geographical locations and time points and extract this information. Try our SweNam system.

Automatic text extraction can be used for:


Responsible for this page: Hercules Dalianis <hercules@nada.kth.se>
Latest change October 10, 2001
Technical support: <webmaster@nada.kth.se>