Analysing the sentiment of people during each phase of COVID-19 Pandemic in India Blog: https://vivekraja98.medium.com/detecting-and-visualizing-twitter-sentiment-during-covid-19-pandemic-using-aws-comprehend-and-c641e1549e2b
The tweets were collected on three bases: 1. Keyword, 2. Location 3. Date range. Only tweets in English were collected for this purpose Keyword: 15 trending keywords related to COVID-19. Location: Chennai, Bangalore, Hyderabad, Delhi, Mumbai, Kolkata. Date Range: Three phases — Pre-lockdown, Lockdown, Unlock. The GetOldTweets3 python library was used to collect the tweets based on the above three criteria. To preprocess the data, NLTK packages were used. To save you from the hassle, I have uploaded the dataset in my Github repo. The techniques used for data collection and preprocessing is explained in the readme file. All dataset contains three features: tweet text, location, phase. https://github.com/Vivek0712/covidtwitterdataset