EXTRAKT's functions are:
it identifies the language of a text, even if terms of different languages are mixed.
These 4 functions analyze the text.
The INDEX functions returns the basic forms, the ANALYZE functions adds logical operators which combine the baisc forms (for instance the parts of a compound word in German). The INDEX2 function splits the text into sentences and carries out a Named Entity Extraction, by this mean identifies persons, proper names (companies, institutions, products), geografical terms and dates. The SUMUP function "summarizes" a text.
The GENERATE function generates all morphological variants to a given basic form (lemma).
The generation can be limited to variants with different stems.
The TRANSLATE function translates the basic forms from one language (source language) into another language (target language). The language pairs are:
- from any of the Extrakt languages into English and German - French, German - Italian and vice versa.
The SYNONYM function combines synonyms, associated terms and derivations. The Syn-Filter combines reduces the list of synonyms by filtering them against a list of predefined terms (usefull with e-shops with a specific vocabulary).
The THESAURUS function combines a given term with hyponyms and hypernyms.
The TRAPHO function transforms a word into a phonetic representation.