Tutorial System in Learning Activities Through Machine Learning-Based Chatbot Applications in Pharmacology Education

Based on the education index issued by the United Nations Development Program and the Global Talent Competitiveness Index, education in Indonesia still needs to be improved through various innovations and infrastructure. One of them, the needs of students in the health sector (including in the field of pharmacology) to obtain learning information easily are still not widely implemented using information technology. Usually, students need to read textbooks and journals to get information that is in accordance with the lesson because there are not many materials about the health sector circulating on the internet, especially in the form of chatbots. One of the solutions that can overcome these problems is to utilize information technology through the use of chatbots in the field of pharmacology learning. In this research, a tutorial system for learning activities is built with a machine learning-based chatbot application. The general architecture of a chatbot consists of NLU components, message processor, dialogue management, action executor, and utterance generator. Based on evaluation, this research have successfully achieved 94.3% accuracy for intent classification and 97.5% accuracy for entity extraction. Majority of the users feel that the chatbot functionality is easy to use and easy to remember. Furthermore, majority of them (90%) thought that tutorial process provided by this chatbot is similar to the real tutorial process provided by human.