Advisory-AI: Fostering Learning Support with an LLM-Powered Dialogue System
EDUCCON 2024 Education Conference, Kahramanmaraş, Türkiye, 27 Eylül 2024, ss.22, (Özet Bildiri)
- Yayın Türü: Bildiri / Özet Bildiri
- Basıldığı Şehir: Kahramanmaraş
- Basıldığı Ülke: Türkiye
- Sayfa Sayıları: ss.22
- Açık Arşiv Koleksiyonu: AVESİS Açık Erişim Koleksiyonu
- Ankara Yıldırım Beyazıt Üniversitesi Adresli: Evet
Özet
ADVISORY-AI: FOSTERING LEARNING SUPPORT WITH AN LLM-POWERED DIALOGUE SYSTEM Hatice Karaaslan1 - Uygar Yüzsüren2 - Özkan Kılıç3 Emirhan Aslankarayiğit1 - Buğra Alptekin Sarı1 - Melisa Boydak1 Gürkan Er1 - Melikşah Bozkurt1 1Ankara Yıldırım Beyazıt University, Ankara, Türkiye. 2İleri Veri Teknolojileri ve Danışmanlık, İstanbul, Türkiye. 3Cisco Systems Inc., Learning & Certification, CA, USA,. ABSTRACT Advisory-AI is a project that offers the potential to restructure learning support by providing students with a readily available, personalized tool to overcome their learning challenges and achieve academic success. It utilizes a Large Language Model (LLM) to deliver personalized learning support through an intentionally reflective dialogue (Kato & Mynard, 2016) structured around Cognitive Behavioral Therapy (CBT) techniques. Learners will engage with the LLM, receiving prompts and guidance to identify and challenge unhelpful thinking styles hindering their progress (based on Beck’s cognitive model). This will be followed by the development of personalized action plans and accountability check points to address these obstacles. The system’s accessibility and scalability will allow for widespread learner support. By measuring learner satisfaction, learning progress and the reduction in self-reported learning challenges, we aim to assess the effectiveness of this approach in empowering learners to achieve academic success. Kato, S. & Mynard, J. (2016). Reflective Dialogue: Advising in Language Learning. Routledge. Keywords: Cognitive Behavior Therapy (CBT), Learning advising, LLM, Personalized learning support