InfoCodex is based on an advanced new technology for text analysis and cross-lingual content recognition (patented in the EU and U.S.A.).
Words or expressions (groups of words such as "European Union", "Enterprise Search Engine" etc.) present in the text are matched against InfoCodex's linguistic database, which comprises more than 3 million words/expressions which are structured into cross-lingual synonym groups. These synonym groups point to a node in a universal taxonomy (ontology) characterizing the meaning of the matched synonym. This language-independent information is then used to extract the content of the document.
Cross-lingual universal knowledge repository
The central multi-lingual knowledge base (linguistic database) is the result of combining and harmonizing knowledge repositories from established works such as the WordNet (Princeton University), EuroVoc, AgriVoc, JuriVoc, the International Classification for Standards (ICS), financial taxonomies and many other sources.