Content
Extraction
Once
the scanned document is classified as
a specific document type, a template is
used
to instantiate a task that extracts the
fields of interest from the document,
based on the data definition in the template.
Extraction Process
The key data from the image is extracted,
and using Fuzzy Matching, the correct customer
data will return from the data base and
populate the customer data attributes.
Fuzzy Matching improves the response-data
extraction substantially to more than 80%
accuracy. 80% of responses need no
manual intervention.
Papyrus FreeForm...
...enables automated processing of unstructured
documents for classification and extraction.
In the classification stage various self-learning
procedures are employed. For the extraction
process the exact position of the individual
fields does not have to be known.
For quick and effective definitions of
the relevant rules and to monitor the self
learning process a powerful FreeForm Designer
is available.
|
