ABSTRACT
COVID-19 outbreak has caused a high number of casualties and is an unprecedented public health emergency. Twitter has emerged as a major platform for public interactions, giving opportunity to researchers for understanding public response to the outbreak. The researchers analyzed 100,000 tweets with hashtags #coronavirus, #coronavirusoutbreak, #coronavirusPandemic, #COVID19, #COVID-19, #epitwitter, #ihavecorona, #StayHomeStaySafe, #TestTraceIsolate. Programming languages such as Python, Google NLP, and NVivo are used for sentiment analysis and thematic analysis. The result showed 29.61% tweets were attached to positive sentiments, 29.49% mixed sentiments, 23.23 % neutral sentiments and 18.069% negative sentiments. Popular keywords include "cases", "home", "people" and "help". We identified "30" such topics and categorized them into "three" themes: Public Health, COVID-19 around the world and Number of Cases/Death. This study shows twitter data and NLP approach can be utilized for studies related to public discussion and sentiments during the COVID-19 outbreak. Real time analysis can help reduce the false messages and increase the efficiency in proving the right guidelines for people.