Self-Medication Chatbot Application Using Natural Language Processing Method

Health is a fundamental need that cannot be ignored by any individual. One common health issue is coughing, which requires appropriate treatment depending on the type—such as antitussives for dry coughs and expectorants or mucolytics for productive coughs. However, many people still engage in irrational self-medication due to a lack of understanding regarding medication and health information. To address this issue, this study aims to design a web-based self-medication chatbot application capable of providing information related to self-treatment by utilizing the Natural Language Processing (NLP) method. The system is built using the Recurrent Neural Network (RNN) approach and the model is trained with augmented data to improve its accuracy in understanding user input. The developed chatbot is capable of recognizing the context of user queries and automatically delivering relevant responses. Test results show that the implemented model achieves high validation accuracy, making it effective in helping users quickly and independently obtain drug-related information. This application is expected to serve as a practical solution to increase public awareness of rational medication use.