A Gamified Learning Tool for Conceptual Modeling with UML Class Diagrams

Conceptual modeling is a crucial part of software engineering's information system design phase, allowing the abstraction of real-world concepts and the translation of complex requirements into a coherent system representation. The Unified Modeling Language (UML) is particularly relevant in this context, with the Class Diagram being a widely recognized and adopted UML diagram type. Teaching and learning conceptual modeling, especially with UML class diagrams, can be challenging due to theoretical and practical complexities. Teachers play a crucial role in conceptual modeling education, but challenges arise when dealing with a large number of students. Also, traditional teaching methods may lack the ability to engage students effectively. In recent years, gamification has emerged as a strategy in software engineering education: it consists of the use of game design elements in non-game contexts to increase motivation and engagement. The main goal of this thesis was to create a gamified educational tool for teaching conceptual modeling with UML class diagrams. First, an in-depth analysis was carried out of the main aspects of gamification and the benefits that its application in various contexts can bring. In particular, the Octalysys framework was analyzed and used as a tool for the application of gamification. After that, the main features and rules of conceptual modeling using UML class diagrams were explored. Based on this, the design and implementation of a gamified web tool for teaching conceptual modeling was carried out. The key feature of the tool is an automatic evaluation system for diagrams created by students. The system performs two different types of analysis: a syntactical one, based on verifying that the syntax rules about UML class diagram modeling are followed, and a semantical one based on a solution that aims to verify the completeness and correctness of the diagram modeled by the students. The gamification mechanics selected and implemented are the following: a system of gaining levels through the acquisition of experience obtainable by completing exercises correctly; an avatar that can be customized with items that can be unlocked by leveling up; a system of immediate feedback after the evaluation of an exercise through highlighting with different colors and description of errors made in a diagram. To evaluate the implemented tool two types of analyses were conducted. The first analysis focused on the automatic diagram evaluation system: a total of 30 diagrams created by students were analyzed by the tool, which generated a list of errors reviewed by a human evaluator to determine the accuracy of the tool’s identification of errors and warnings. The analysis revealed that the system is able to correctly detect a very high percentage of violations, in a manner very similar to the evaluation of a human being. For the second analysis, a score expressing the completeness of the gamified environment was calculated using the method provided by the Octalysis Framework. The analysis showed that the gamified experience appears to have a balance between the different core drives defined in the framework. However, it becomes apparent that some core drives are not well-represented at present. For this reason, future development should expand the tool’s features with other gamified mechanics that are suited for long-term usage, such as competition mechanisms like leaderboards and quest-line mechanics based on the exercises.