Publication Details
Abstract
The rapid advancement of technology in education has created new opportunities to improve the effectiveness and efficiency of assessment practices. This paper explores the development of software designed to automatically generate control tests and ticket questions based on content materials in higher education. The proposed system uses natural language processing (NLP) techniques and machine learning algorithms to analyze educational content and generate relevant, accurate, and diverse test questions. The system's performance was evaluated in terms of question quality, cognitive coverage, efficiency, and user satisfaction. Results indicated that the software achieved high levels of clarity (88%), relevance (91%), and correctness (89%) in the generated questions. However, its ability to generate higher-order questions (e.g., analyzing and evaluating) was less pronounced. The system also demonstrated significant time-saving benefits, reducing the question creation time from 120 minutes (manual) to just 15 minutes (automated). User feedback was highly positive, with 92% of users reporting that the system saved time and 85% expressing interest in using it for future courses. Despite its successes, the system requires further development to improve its ability to generate complex questions and handle diverse content types. This paper contributes to the ongoing discussion on the integration of automation in educational assessment practices.