Development of Intelligent Telegram Chatbot Using Natural Language Processing

Intelligent chatbots have been gaining interest in the past years due to the advance of artificial intelligence algorithms. As a result, many studies have been conducted on emotional transition and dialog structures. One of the benefits is on the medical applications, in which psychological assessment, clinical counseling, autism diagnostics, and advanced cognitive models could be provided. On the other hand, chatbot knowledge generally comes from a web-based information repository, in which the information is reliable, but it is rather not versatile as it does not contain emotions. The objective of this paper is to develop an intelligent chatbot using natural language processing and Telegram API. First, text processing using Telegram API on Python was developed. Next, emotion recognition was performed on the recorded chats. The appropriate response is then sent to the user. Results showed that our Telegram chatbot could interact smoothly with the users and identify the user's emotions.