SERdistiller: Automated Document Capture and Data Collection

The next wave in classifying and indexing unstructured information with minimum user intervention. - Imaging and Document Solutions Magazine

SERdistiller is a software solution that uses neural network techniques to extract information from documents and manage content without tedious programming or data entry. It replaces the manual effort required to classify and index every document that goes into an imaging and knowledge management system. The solution literally mimics human learning by classifying documents based on similar content, which the system has seen before. SERdistiller classifies the full range of document types from completely unstructured to fully structured documents. It also extracts text and numerical fields for index population.

Classifies Structured & Unstructured Content

Structured Documents -Fully structured documents, such as tax forms, have all data fields in fixed horizontal and vertical positions.
Loosely Structured Documents -Many document types, such as received invoices, and delivery notes, have a consistent set of fields, which are at variable locations within the document. This type of form is termed loosely structured.
Unstructured Documents -examples include reports, memos, letters and magazines, in which there are no consistent set of fields or field locations within the document.