Sentiment analysis on covid-19 awareness conversations using CNN

Abstract

Natural Language Processing (NLP) has become one of the most significant technologies of the 4th Industrial Revolution and a well-liked field of AI with the rise of voice interfaces and chatbots. The discipline of NLP has produced a wide range of beneficial applications, which are rapidly expanding. They range from basic to intricate. One is sentiment analysis. Two years back, the whole world was hit by a pandemic, the Covid-19. Many people lost their lives, many people lost their jobs. Schools, colleges, universities, offices all were shut. The whole world came to a standstill. Even today, the effect of Covid has not been erased completely. Daily new cases are being reported. Standing in this situation, if we could get the help of an advisor who can give the necessary solutions would have been a great help. This could also help in spreading awareness about the pandemic. The goal is to create an automated advisor using NLP. A chatbot can be trained over conversations on Covid-19. For training the chatbot, a large number of conversations are required. Along with it, intent words and sentiments are also required to understand the intention of the user and the depth of the situation. We have developed a dataset where two users (User1 and User2) will take turns to chat among themselves regarding Covid-19. However, we have employed several machine learning and one deep learning model like CNN to explore the insights of sentiment analysis on COVID discussions.