/
Business Data
Work Phone: 
0049 89 69 3333 313
Personal Bio

Since 2005, Michael Fuchs has served in different roles as the technical (product marketing) “middleman” between market and customer requirements, the ABBYY Europe Sales Teams and product development groups in the ABBYY headquarter. The focus of his work is on toolkits and cloud services for developers, integrators and ISVs. In his new role as a Technology Evangelist he supports ABBYY’s business on a worldwide level and continues to promote ABBYY’s innovative, upcoming semantic technologies as well as the document recognition technologies.

Presentation title: 
Spotlight presentations
Presentation description: 
ABBYY is leading software vendor known for its high quality optical text recognition technology products and solutions. This talk presents ABBYY’s new linguistic technology platform “Compreno”. ABBYY Compreno processes natural language text in a new way, just as a human brain would. ABBYY’s approach combines intense linguistic analysis of natural text with a language-independent universal semantic hierarchy. This makes it possible to detect entities, facts and the relationships between different objects in unstructured natural language documents and to provide the semantic meaning of the words used in the texts. This approach generates a lot of meta-data about the processed documents which enables computers to understand the meaning of text. The embedded linguistic intelligence does not rely on statistical training. Therefore ABBYY Compreno technology can be used in scenarios where the volume of text/documents is too low for statistical analysis. This can be the case when only short text snippets or a small/medium document volume have to be analysed, like contracts or accident reports. The results can be made available as “raw” linguistic XML or as “condensed” RDF/XML. The talk gives an overview on the results of over 15 years scientific R&D with up to 250 linguists involved. Currently the technology is on its way to leave the labs. ABBYY developed a processing backend that applies the linguistic analysis and fact-relationship extraction. The API is now looking for real live projects (and partners).