Development of a Virtual Mentor Integrated with Retrieval-Augmented Generation Artificial Intelligence for Project-Based Learning

Abstrak

Project-Based Learning (PjBL) is a learning model that can enhance the quality of higher education, particularly in developing critical thinking, creativity, and collaboration skills. However, implementing PjBL often faces challenges such as limited resources and the need for intensive guidance from lecturers. To overcome these challenges, Artificial Intelligence (AI) technology offers great potential, although traditional AI systems often provide responses that are less relevant to the context of the learning material. The Retrieval-Augmented Generation (RAG) technique in AI can serve as a solution, enabling the system to generate more accurate and contextually relevant responses. By utilizing data sources such as course materials, RAG can enhance the relevance of AI responses in supporting project-based learning. It is expected that developing an AI-based virtual mentor using the RAG approach can optimize students’ PjBL experiences. Specifically, this virtual mentor is designed to provide contextual guidance, help students overcome project-related challenges, and foster independent learning, thereby improving the quality and effectiveness of PjBL in higher education.

Identifier

• OpenAlex: https://openalex.org/W7135179494
• DOI: https://doi.org/10.32664/smatika.v16i01.2258