Implementation of Generative Pre-Trained Transformer 3 Classify-Text in Determining Thesis Supervisor

Abstract

One of the requirements for graduating from the undergraduate level for universities in Indonesia is writing a final project or thesis. In order to graduate, of course, it is greatly influenced by the desire and strong spirit of the students and also the guidance of the supervisor. In determining the supervising lecturer, special attention must be paid to the field. Usually the selection of lecturers for thesis supervisors is determined by the study program through a meeting of lecturers in order to determine which lecturers are considered according to the title of the student and in accordance with the research of the supervisor. However, this method is a bit inconvenient and also quite time-consuming considering the number of students is more than a hundred and continues to grow every year. In this study, the thesis supervisor was classified based on the title proposed by the student. The methodology that will be used in this research is the Cross-Industry Standard Process for Data Mining (CRISP-DM) methodology whose stages are: Business Understanding, Data Understanding, Data Preparation, Modeling, Evaluation, and deployment, as well as using Generative Pre-Technology. trained Transformers 3 (GPT-3)

GS Cited Analysis