Automatic Document Extraction - Enrich Business Processes with Data

After incoming documents are classified and assigned to a business process, as part of the automated data extra extraction, those customer and product related features are read that are relevant for the further processing of the business process. These could be key words (customer or contract number) or identification features that reconcile with CRM or ERP systems. 

Here the ITyX technology for automated document extraction offers two approaches. On the one hand the ITyX technology has processes for a rule-based reading-out of information from documents. On the other hand ITyX offers a unique process of self-learning data extraction in unstructured documents. Through 'iTAG', data fields are learnt that are used in each business process at a later stage, or have to be highlighted to staff during the transaction processing. Through the FuzzyMatcher fuzzy search processes for intelligent reconciliation of transaction data with legacy systems can be executed.

The use of automated data extraction is - as in all other capturing methods - controlled through the graphic process designer. Via the designer, SML import and exportof data from the transaction and/or the transaction process is carried out. 

ITyX Technology for Document Extraction includes eg:

• Graphic processes (eg position related data in forms)
• Linguistic processes (eg content related data in running text)
• Rule based process of document extraction
• Self-learning extraction of data through analysis of observed process steps
• Capture of field relations (validation and conversion processes)