Preparation For Science Baccalaureate with Artificial Intelligence Application: The Stewdy Example
18th International Congress On Social, Humanities, Administrative, and Educational Sciences In a Changing World, Ürgenç, Uzbekistan, 10 - 12 March 2026, pp.450-464, (Full Text)
- Publication Type: Conference Paper / Full Text
- City: Ürgenç
- Country: Uzbekistan
- Page Numbers: pp.450-464
- Ondokuz Mayıs University Affiliated: Yes
Abstract
Artificial intelligence can be used in various ways in the preparation process for both local and national exams. While these systems offer personalized learning materials by analyzing students’ knowledge levels and learning styles, the adaptive learning system focuses on learning analytics and data mining techniques. On the one hand, it helps teachers prepare lesson plans for teachers and on the other hand, it designs student specific learning materials. When studies conducted at the international level, each country develops its own artificial intelligence systems due to the differences in university entrance exam systems. The characteristics of the exam language have made it compulsory for some countries to design artificial intelligence applications suitable for their own exam systems. In France, two-thirds of students receive support in preparation for the baccalaureate exam, which is called the high school graduation exam. Artificial intelligence systems are among the most important materials supported. The aim of this study is to examine the Stewdy artificial intelligence application designed for the preparation of the baccalaureate exam in the field of science in France. For this purpose, a case study from qualitative research methods was preferred. While the interface and features of the Stewdy application were analyzed descriptively, the answers given by the chatbot were analyzed as a case study. As a result of the descriptive analysis, it was observed that the artificial intelligence application helped in four main topics. These are the exam syllabus, a specific outcome, testing with exercises, and research on the image that can be uploaded. Since the baccalaureate exam contains open-ended questions, the use of both outcomes and formulas is important. Questions in physics and chemistry need to be context based and use different skills, from naming a compound to drawing its chemical structure, from reading a graph to drawing a ray. In the exercises, it was observed that this system can provide different levels of feedback through the chatbot. In addition, the fact that there is a calculator allowed to be used in interface specific exams and that all mathematical formulas that can be specified on paper are written on the screen is understood to be at a level that can meet the needs. In the light of the findings, it was concluded that the Stewdy was able to provide personalized feedback based on questions, explain in depth with different examples each time, and answer the questions uploaded as images correctly.
Keywords: Artificial Intelligence, Science Education, Baccalaureate, Exam Preparation, Stewdy.
* This study was supported by TÜBİTAK 2219 Post-Doctoral Research Program