Fig. 2From: Natural language processing for automatic evaluation of free-text answers — a feasibility study based on the European Diploma in Radiology examinationProcess diagram illustrating our approach. Based on the task description, the correct answer and marking description concepts with synonyms and training examples are modeled (step 1 as described in the text, blue-dotted box). Based on concepts and synonyms, a rule-based concept detection is learned (step 2 as described in the text, yellow-dotted box), and at the same time, a decision tree is learned on the training examples (step 3 in the text, red-dotted box). Finally, the NLP can be run (step 4 in the text, green-dotted box)Back to article page