Automatically classifying texts and extracting information is the key to customer-friendly and fast bureaucratic processes. What previously could only be done by humans in painstaking detail work, we can partially and fully automate.

Applications are among others:

  • Classification of PDF documents, e.g. into classes such as "Rental Contract", "Offer" or "Invoice" with high recognition rate.
  • Topic recognition in documents. Here documents are divided into their structured components.
  • Extraction of metadata for storage documents
  • Information extraction from invoices and other structured documents
  • Automatic XML tagging of texts

The applications are constantly growing.

portamis has its own KI4Text library written in Java, which is constantly extended and used in such projects.