DEVELOPMENT OF INSTRUCTOR MODULE IN AN INTELLIGENT TUTORING SYSTEM FOR ENGLISH LANGUAGE USING LARGE LANGUAGE MODEL

• Main Author: Ho, Vincent
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/82405
• Institution: Institut Teknologi Bandung

Based on research conducted by Education First, Indonesia has a low proficiency in English. This situation is worsened by the lack of teaching resources and the unequal distribution of education across the country. Consequently, Indonesia struggles to actively participate in the global environment. To address these issues, there is an opportunity to develop an Intelligent Tutoring System (ITS) for English language learning using Large Language Models (LLM). ITS can reduce the time teachers spend on the teaching process, thereby optimizing the existing human resources in education. In this final project research, the instructor module of an ITS for English language learning has been developed. The implementation of the instructor module is crucial for supporting an effective and efficient teaching process. The instructor module must accommodate all the needs of teachers in conducting their teaching activities, which led to the implementation for several key features, including course management, material management, exercise management, student progress tracking, and chatbot. The selection of these features was based on literature review and consultation with industry experts. The instructor module was developed using TypeScript and the React library for the frontend, and Python and the FastAPI framework for the backend. The LLM is utilized in the form of a chatbot that acts as a personal teaching assistant. Testing results indicate that the ITS instructor module for English language learning using LLM successfully meets the defined functional requirements. The average SUS score is 92.1 and the SEQ score is 6.81, indicating that the application is both usable and easy to use.