DOKuStar Professional

Document Interpretation

DOKuStar Professional is software for document analysis that automatically classifies digital documents, extracts data and delivers structured data (e.g. TIFF->XML). The central processing modules in DOKuStar Professional are DOKuStar Extraction, DOKuStar Adaptive Recognition Technology (ART) and Adaptive Classification Technology (ACT).

DOKuStar Professional Architecture

DOKuStar Professional works with configurable processing chains or pipelines. In a processing chain, a document passes through one module after another. The system calling DOKuStar views it as a component (sub-system).

DOKuStar Professional offers four significant enhancements above and beyond the functions offered by DOKuStar Extraction:

In addition to rule-based classification and data extraction, DOKuStar Professional offers a learning procedure. Referred to as ART (Adaptive Recognition Technology), this patented procedure allows users to indicate to the system where the required information is located in a document by moving the mouse pointer over sample documents. Afterwards, ART can find this information in similar documents. This new procedure significantly smoothes the system optimisation process and noticeably increases recognition performance within the applications. The rule-based classification in DOKuStar Professional can be supplemented with the self-learning process, Adaptive Classification Technology (ACT). The ACT module features a automated, content-based classification of unstructured documents. With the high-performance and easy to use administration tool the recognition accuracy can be optimised significantly.

PDF as input source

Adobe's "Portable Document Format" is becoming more and more important in the world of documents. The PDF/A standard now makes archived PDF documents future-proof, ensuring view ability for the future. This makes PDF a very real alternative to TIFF in archiving scanned documents.

DOKuStar Professional processes multiple page documents. The software can even support complex document structures with sub-documents. As a result, even extremely demanding mail inbox solutions can be quickly displayed.

The recognition functions can be quickly adapted to suit specific project requirements using the various programming interfaces. An off-the-shelf system, by contrast, cannot always meet the needs of demanding projects and users. Company databases or flow controls, which depend on the content of a document, can be accessed via .Net or COM interfaces.

Load distribution over a server cluster ensures that DOKuStar Professional always uses the available processing power. The server interface is just as simple to use as the interface of the recognition software itself. This ensures that DOKuStar Professional can be directly integrated into any enterprise application.

Related Documents

DOKuStar Professional Brochure (English - PDF)
DOKuStar Professional Brochure (Deutsch - PDF)
DOKuStar Professional Brochure (Español - PDF)