Recommendation of new semi-automated approach of decision tree diagram

Assessment of tree diagrams is time-consuming, especially if the diagrams are paper-based. This is because if the student creates a diagram without a clear understanding of the scenario text, this makes the diagrams unmanageable, especially if it is presented on A4 paper. This paper investigates and analyses the case study of the decision tree. Decision tree (DT) diagrams are one of the most commonly used data mining methods [1], [2]. Marking decision trees has many difficulties; one of these is represented by the subjectivity of the marking process. The subjective nature of decision tree assessment leads to variations in marks given by different assessors, which is seen by students as a great source of unfairness. For this reason, a new approach to semi-automated assessment systems is to be developed. This approach might enable the use of computers to support the assessment process of students' diagrammatic solutions, especially relating to decision tree diagrams. This paper summarises and analyses the case study of the decision tree and provides the requirements of the new approach.