EDUBOT - DEVELOPING KEY COMPETENCIES THROUGH BLENDED-LEARNING METHODOLOGY BASED ON AI-SUPPORTED CHATBOT TECHNOLOGY

One of the biggest challenges in traditional education is ensuring learning that matches the individual abilities of students There is now clear evidence that school practices that provide the same content to all children and use the same methods are not effective. In addition, appropriate differentiation, taking into account the individual needs and characteristics of each student, is difficult to achieve using traditional teaching methods in often large class sizes and requires a significant investment of time and energy by teachers. School opportunities are also affected by other socio-cultural factors, with children from disadvantaged backgrounds not having access to quality education, being more at risk of early school leaving and having low levels of basic skills. The project has provided an innovative solution to these needs by developing a differentiated education solution supported by AI solutions, including a blended learning methodology, an e-learning support system and a digital content. This solution focuses on personal learning needs, supporting adaptive learning processes, students are presented with challenges that are challenging but not frustrating.

The aim of the project was to create an innovative, effective and free blended learning methodology supported by digital solutions that apply differentiated teaching in STEM education. Our aim was primarily to provide support for students with fewer opportunities, to promote equal access to education, including equal access to higher levels of education, and to contribute to combating early school leaving and developing basic competences by providing adaptive, personalised digital learning routes and small group tutoring to identify and address competence gaps. At the same time, we have sought to ensure that the project outcomes allow for a wide range of adaptability across different areas of education. Specific objectives: - to create an e-learning environment supported by AI solutions - Developing a blended e-learning methodology - Developing experimental STEM digital content - Testing the results in a pilot session - Creating an adaptation toolset to to make the results developed available to schools and institutions working with disadvantaged students - Wide dissemination of results

A blended methodology was developed, which consists of two phases: first, students learn with the digital learning material on the e-learning platform, and then teachers tutor and develop students in cluster consultations. 4 innovative digital STEM contentpools with interactive and multimedia elements have been developed, using a digital methodology based on AI-driven adaptive learning routes, which focus on differentiated learning and allow for personalised learning, with students following different learning routes according to their abilities and current knowledge. In the pilot session, the results were tested in a school environment, involving 4 countries, 11 institutions, 20 teachers and 569 students. In 3 countries, maths exam preparation modules were implemented to prepare students for the secondary school entrance exams in Hungary, the 8th grade National Evaluation Exam in Romania and the 8th grade exam in Polish primary school, and in Slovakia students were working on practice-oriented knowledge in computer science. The overall results show an improvement in performance of 20.38%. Intensive networking and dissemination activities have been undertaken to ensure impact and sustainability.

One of the main outputs of the project is the Blended Learning Methodology, which is a summary of all the project outputs and the main tool for adaptation. It is available as an e-book and a printable manual in 5 languages. Blended learning methodology: combines digital learning with online small group tutoring, combining the benefits of AI-based and teacher-led learning. The digital learning materials are processed in the e-learning environment, followed by further tutoring in homogeneous small groups resulting from performance-based group segmentation. The Edubot e-learning support system has been implemented.: the EDUBOT Tool (for teachers), the EDUBOT App (for students) and the EDUBOT Assistant. 4 innovative digital content pools: content that can be navigated along adaptive learning routes, with tasks of varying difficulty levels, helping exercises and explanations to help students. In 3 countries, mathematics topics for the 8th grade or secondary school entrance exams were developed, while in one country, a curriculum in computer science was developed, A total of 28 maths and 9 IT subjects have been created. All results are free to use and adapt for all stakeholders, not only in STEM.