Chatbot development is a fascinating and challenging field that combines software engineering, natural language processing, and machine learning. I have built a health chatbot prototype called “CakapSihat”, as part of my Master’s thesis in Software Engineering. It aims to provide health information and advice to users in a friendly and conversational way. It can answer questions about common health topics, such as diseases, fitness, and healthy lifestyle.

I have used JavaScript as the main programming language for both the frontend and the backend of the chatbot. This made the development process easier and more consistent, as I did not have to switch between different languages or frameworks. For the backend framework, I used NodeJs, which is a fast and scalable platform that allows me to create web applications using JavaScript. NodeJs handles the communication between the frontend and the backend, as well as the integration with Google Dialogflow and MongoDB.

For the frontend framework, I used React, which is a powerful and popular library that allows me to create interactive and dynamic user interfaces using JavaScript. React renders the chatbot’s user interface, which consists of three webpages and a chat window. The chat window was designed using Materialize, a modern and responsive CSS framework.

For the natural language processing platform, I used Google Dialogflow, which is a cloud-based service that provides a simple and intuitive way to create and manage conversational agents. Dialogflow handles the chatbot’s intelligence, such as understanding the user’s input, generating the appropriate response, and maintaining the context of the conversation.

For the database, I used MongoDB, which is a flexible and scalable document-based database that allows me to store and retrieve data in JSON format. MongoDB stores the chatbot’s knowledge base, which consists of health-related information and resources.

Due to the time constraint of my thesis, I was only able to give limited training to the chatbot. As a result, the chatbot demonstrates the functionality of a working chatbot, but the accuracy of its interpretation is very precise. However, I believe that with more training and refinement, the chatbot can become a useful and reliable tool for health education and promotion.