Abstract:

 Sentiment analysis is a process that uses natural language processing and machine learning to extract and identify the emotion associated with a text or an image. In the literature, we find different approaches to sentiment analysis that predominantly fall into two categories: machine-learning-based and lexical-based. When using machine learning for sentiment analysis, usually a classifier is required. When applying lexical-based methods, a list of known words, with each word having an associated sentiment, is used. The aim of the paper is to analyze the opinion of Instagram users towards the lockdown and social distancing caused by the COVID-19 pandemic, which spread rapidly worldwide at the beginning of 2020, with the use of various sentiment analysis tools. All the data collected for this study is public data gathered with the help of a tool called Phantombuster and is composed of over 15,000 posts. The posts were selected using several relevant hashtags and for the actual sentiment analysis process, Google Cloud was used. The collection and interpretation of the data are described in depth in this paper, providing suitable examples and possible explanations for the obtained result. The study might have importance for decision makers, but also for researchers, who have to choose or combine several sentiment analysis instruments in order to obtain meaningful results in their field of study.