Development of chatbot on Web application

Abstract

Nowadays, AI can replace some position in the society due to their repetitive work property. And the job of waiter and waitress in restaurant is forecasting that will be replaced by AI in the next decades. The objective of this project is using deep neural network to train a chatbot to act a waiter in a restaurant. This bot can help with taking order from customer, giving recommendation to the customer, answering other pre-defined questions, which is so-called intent. Involving 2 main technique in the field of Natural Language Processing in AI – Intent classification and slot filling, this project is projected into a web application. Making use of modern neural network model for natural language processing (NLP) – Bert embedding, Attention based LSTM, this bot can achieve an excellent ability of understanding natural language. It is hoped that this project can be using in our daily life and contribute to our society.