DDC automates the recognition of semi-structured documents, such as invoices, sales orders, delivery notes, etc.
How it works
Assuming the document type is accurately identified, DDC automatically extracts all or part of the document key fields, such as the invoice or sales order number or the lines of details.
DDC ensures the accuracy and the validity of all data extracted by using multiple mechanisms such as:
Consistency: The values are consistent with one or more other values (e.g. unit price * quantity = extended price)
Matching: Extracted data exists in your ERP database or in a third party repository. (e.g. a supplier or a customer is already known by the ERP system)
Validation of the extracted data is integrated into a complete workflow process.
A validation form is visible in a web interface which displays extracted data for validation, improvement or correction.
On the left is the extracted data, on the right is the document image.
The validation workflow supports out of the office management.
Teaching (content analysis improvement)
Because it is impossible to take into account the full variety of document variations, the system can be taught to recognise certain types of documents with more accuracy.
The process is constantly being updated to be more complete and effective; the system gets more and more intelligent overtime.
Teaching is available only to users with the requisite access rights (power users).