DEVELOPMENT OF STUDENT MODULE IN INTELLIGENCE TUTORING SYSTEM FOR ENGLISH LEARNING USING LARGE LANGUAGE MODEL

• Main Author: Christian Siregar, Vincent
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/83743
• Institution: Institut Teknologi Bandung

Education plays a crucial role in the advancement of a country, especially in facing the challenges of globalization that demand international competition. English, as an international language, is particularly important in this context. However, the level of English proficiency in Indonesia, ranked 81st out of 111 countries with a score below the regional average in Asia, remains low. Information technology, through applications like Intelligent Tutoring Systems (ITS), offers a solution to this issue by providing personalized and efficient English language learning. ITS, which integrates student modules, tutor modules, and domain modules, can help overcome the limitations of the number of teachers and access to education by enabling students to receive materials and exercises that are tailored to their abilities. In this final project, the student module of the ITS component was developed. The student module interacts directly with learners who will use the application. The application was designed to be easily accessible to students, so it is available in both desktop and mobile formats. Several features were implemented in this final project, including authentication, courses, materials, exercises, student progress tracking, and a chatbot. The chatbot feature utilizes a Large Language Model (LLM) to serve as a tutor assistant in helping students learn English. The application consists of a frontend and backend. The frontend application is developed using the Typescript programming language and the ReactJS framework. The backend application uses the Python programming language and the FastAPI framework. An evaluation was conducted to determine whether the application is functional. The evaluation was performed using usability testing methods through interviews. The evaluation results showed that the application received a score of 6.78 for SEQ testing and a score of 97.5 for SUS testing. These results demonstrate that the application is suitable for use.