Generating a structure of a PDF-document

A computer program product for generating a structure of a PDF-document, wherein the PDF-document comprises elements and pages, the computer program product comprising one or more computer-readable storage media, and program instructions collectively stored on the one or more computer-readable storage media, the program instructions comprising:

• instructions for detecting document cells of the PDF-document dependent on commands of a page description language for printing the elements of the PDF-document, wherein the PDF-document comprises the commands, wherein the detecting of the document cells includes detecting a respective position of a respective document cell;
• instructions for determining parts of the PDF-document dependent on the PDF-document by a machine learning module (ML module), wherein the determining of the parts includes determining a respective position of a respective part on the pages, wherein the respective part is represented by a respective box on the pages, the determining of the respective part of the parts comprising associating a respective portion of the elements of the PDF-document that is covered by the respective box with the respective part;
• instructions for assigning a respective label to the respective part;
• instructions for using a symbolic artificial intelligence module (AI module), wherein the using of the symbolic AI module comprises applying rules of the symbolic AI module for reconciling the document cells with the parts, wherein the applying of the rules includes reconciling the respective position of the respective document cell with the respective position of the respective part;
• and
• instructions for generating and labelling elements of the structure of the PDF-document dependent on a result of the reconciling and dependent on the respective label to the respective part.

A computer system for generating a structure of a PDF-document, wherein the PDF-document comprises elements and pages, the computer system being configured for:

• detecting document cells of the PDF-document dependent on commands of a page description language for printing the elements of the PDF-document, wherein the PDF-document comprises the commands, wherein the detecting of the document cells includes detecting a respective position of a respective document cell;
• determining parts of the PDF-document dependent on the PDF-document by a machine learning module (ML module), wherein the determining of the parts includes determining a respective position of a respective part on the pages, wherein the respective part is represented by a respective box on the pages, the determining of the respective part of the parts comprising associating a respective portion of the elements of the PDF-document that is covered by the respective box with the respective part;
• assigning a respective label to the respective part;
• using a symbolic artificial intelligence module (AI module), wherein the using of the symbolic AI module comprises applying rules of the symbolic AI module for reconciling the document cells with the parts, wherein the applying of the rules includes reconciling the respective position of the respective document cell with the respective position of the respective part;
• and
• generating and labelling elements of the structure of the PDF-document dependent on a result of the reconciling and dependent on the respective label to the respective part.