Akbar Iskandar, Badrun Kartowagiran, Haryanto, Erwin Gatot Amiruddin, Ansar Said
New student admissions at various universities experience a lack of direct interaction in terms of providing information automatically, especially STMIK AKBA. This study aims to build a chatbot application by applying natural language processing, machine learning and finding out how well the response accuracy of the chatbot is. This research was designed by utilizing the tools from google dialogueflow, and the results obtained from this study resulted in a chatbot-based communication media. Furthermore, based on the measurement results of the Usability aspect as a whole, it produces a feasible value for use, while the level of accuracy of the method used in the test results for the threshold value for machine learning is 0.6, while the results of the automatic correction test on user text are 80%. © 2023 American Institute of Physics Inc.. All rights reserved.
Department of Education Research and Evaluation, Yogyakarta State University, Yogyakarta, Indonesia; Department of Informatics Engineering, STMIK AKBA, Makassar, Indonesia; Cendekiawan Inovasi Digital Indonesia (CEDDI), Makassar, Indonesia
