Abstract:

Abstract: Generative AI presents both challenges and opportunities for educators. This paper explores its potential for automating the creation of programming exercises designed for automated assessment. Traditionally, creating these exercises is a time-intensive and error-prone task that involves developing exercise statements, solutions, and test cases. This ongoing research analyzes the capabilities of the OpenAI GPT API to automatically create these components. An experiment using the OpenAI GPT API to automatically create 120 programming exercises produced interesting results, such as the difficulties encountered in generating valid JSON formats and creating matching test cases for solution code. Learning from this experiment, an enhanced feature was developed to assist teachers in creating programming exercises and was integrated into Agni, a virtual learning environment (VLE). Despite the challenges in generating entirely correct programming exercises, this approach shows potential for reducing the time required to create exercises, thus significantly aiding teachers. The evaluation of this approach, comparing the efficiency and usefulness of using the OpenAI GPT API or authoring the exercises oneself, is in progress.

Abstract: This thesis outlines the development of the teacher's User Interface (UI) and the back-end system for Agni, a web playground for learning JavaScript. The goal of this work is to deliver a good User Experience (UX) to teachers through an intuitive and user-friendly UI containing useful functionalities. Therefore, the focus shifts from the more commonly explored research on student experiences to the teacher. To realize this goal, the initial hypothesis was to use a headless Content Management System (CMS), Strapi. However, it was evaluated as unintuitive and time-consuming. Consequently, the development of a new UI became crucial, employing the headless CMS just for the Application Interface (API) and back-end data storage. The primary concept for the developed UI was to use the existing Agni student UI and make it editable. Also, several innovative features were integrated, including the integration of an external exercise repository, namely AuthorKit, the statistical analysis of student engagement, and content sequencing. Further, an experiment on generating programming exercises using GPT was conducted, and GPT was integrated into the system to assist teachers in the process of creating exercises. Finally, the evaluation of the developed UI revealed significant enhancements compared to the Strapi assessments, with an overall positive review. Evaluators primarily highlighted the intuitive and simplistic nature of the UI, thereby affirming the successful realization of the principal objective: to create a teacher's UI that is both intuitive and user-friendly.

Abstract: The paper discusses an ongoing project that aims to enhance the UX of teachers while using e-learning systems. Specifically, the project focuses on developing the teacher’s user interface (UI) for Agni, a web-based code playground for learning JavaScript. The goal is to design an intuitive UI with valuable features that will encourage more teachers to use the system. To achieve this goal, the paper explores the use of a headless Content Management System (CMS) called Strapi. The primary research question the paper seeks to answer is whether a headless CMS, specifically Strapi, can provide a good UX to teachers. A usability evaluation of the built-in Strapi UI for content creation and management reveals it to be generally consistent and user-friendly but challenging and unintuitive to create courses with programming exercises. As a result, the decision was made to develop a new teacher’s UI based on the existing Agni UI for students in an editable version. Once the development is complete, a new usability evaluation of the fully developed teacher’s UI will be conducted with the Strapi UI evaluation as a baseline for comparison.