Structure data

Companies, authorities and research institutes are confronted more than ever with an ever increasing flood of information. There are over 20 million scientific professional articles worldwide, for example, on specific fields and every year there are approx. 500,000 more. In the other areas the need for specific information extraction is increasing as well: commercial enterprises assimilate several hundred thousand orders in their enterprise resource planning (ERP) daily, hospitals register and process several thousand medical reports per year or lawyers analyze emails and server data terabyte-wise for company takeovers.

Free texts become structured data

In order to be able to work with these enormous amounts of data, these must first be compiled, classified and assigned with index terms, as most of them exist solely as unstructured free texts, which can neither be systematically searched nor evaluated. Until now, manual classification and indexing were required which has a long tradition, most notably in the library field. Today, more automated, software-based processes prevail.

Today teh quality of these methods can be closely compared with this of manual processing - and that with considerably less time expenditure. In addition to the mere structuring of data, they also enable its quick and detailed evaluation. The methods of information retrieval and text mining are realized within the Averbis Extraction Platform, which are able to achieve an effective gaining of insights. Here, contents go through an inter-lingual search, are inter-connected and evaluated throughout variously structured documents