Sentiment Analysis of COVID-19 Nationwide Lockdown effect in India
Nida Afroz, Moinak Boral, Vijeta Sharma, Manjari Gupta
Abstract
Nowadays, Sentiment Analysis has become an active research area due to the availability of many opinionated data through increased activity in Blogging, Tagging, Podcasting, social networking sites, RSS feeds, and Social Bookmarking. In the present situation, the whole world is facing the crisis of the COVID-19 pandemic. Particularly, let's talk about nationwide lockdown in India to control the spread of COVID-19. The government relies on social media to observe people's aviews on their policies during the lockdown. In this paper, Twitter data has been used for Sentiment Analysis, which focus on people opinion during the COVID-19 nationwide Lockdown effect in India. Different keywords data was collected on various dates between March 25, 2020, to June 09, 2020. This research work is an application of the real-time TextBlob sentiment analyzer tool built based on the Natural Language Toolkit (NLTK). Relevant keyword tweets were extracted by tweeter API. Then a model was trained to classify the result on a specific opinion. This NLPbased sentiment analysis model is ideal for analyzing the emotions while tested with seven primary keywords: Lockdown1.0, Migrant Workers, Indian Economic, ICMR, Lockdown5.0, Medical Facilities, and Police. The result shows that Lockdown 1.0 got the most positive sentiments, followed by ICMR and Medical Facility.