Print system, print setting prediction method, and non-transitory computer-readable storage medium

A print setting prediction method in a print system for providing print data to a printer across a network, the method comprising:

• accepting document data as a print target and managing the document data by a managing unit;
• issuing, when a print request for document data managed by the managing unit is accepted, print data based on print setting indicated by the print request;
• causing an accumulation unit to accumulate, as collected data, print setting used for print data issued by the issuing and document attribute information indicated by document data used for the print data in association with each other;
• generating, from the collected data accumulated in the accumulation unit, training data including a pair of input data containing the document attribute information and ground truth data containing the print setting;
• performing, by machine learning, generation of a learning model to be used to predict print setting, by using the training data generated in the generating;
• predicting, when accepting a prediction request, print setting by using the learning model generated in the performing and by using document attribute information indicated by the prediction request as input data;
• and
• providing the print setting predicted in the predicting to a request source of the prediction request.

A non-transitory computer-readable storage medium storing a program for causing a computer to function as:

• a managing unit configured to accept and manage document data as a print target;
• an issuing unit configured to issue, when a print request for document data managed by the managing unit is accepted, print data based on print setting indicated by the print request;
• an accumulation unit configured to accumulate, as collected data, print setting used for print data issued by the issuing unit and document attribute information indicated by document data used for the print data in association with each other;
• a generation unit configured to generate, from the collected data accumulated in the accumulation unit, training data including a pair of input data containing the document attribute information and ground truth data containing the print setting;
• a processing unit configured to perform, by machine learning, generation of a learning model to be used to predict print setting, by using the training data generated by the generation unit;
• a prediction unit configured to predict, when accepting a prediction request, print setting by using the learning model generated by the processing unit and by using document attribute information indicated by the prediction request as input data;
• and
• a providing unit configured to provide the print setting predicted by the prediction unit to a request source of the prediction request.