COVID-19 Tweets Textual Analytics Using Machine Learning Classification for Fear Sentiment

2020 International Journal of Advanced Trends in Computer Science and Engineering  
The volume of COVID 19 microblogging messages is increasing exponentially with the popularity of COVID 19 microblogging services. With the huge number of messages seem in user interfaces, it obstruct user accessibility to useful information hide in disorganized, incomplete and unstructured text message. In order to increase user accessibility it present to aggregator related COVID-19 microblogging message into the cluster and automatically allocate them semantically meaningful labels. However,
more » ... typical features of COVID-19 microblogging messages is that they are much shorter than standard text document these messages provide insufficient terms of information for capturing semantic associations providing solution to this problem it suggest a novel framework for organizing unstructured COVID-19 microblogging messages by transforming them into semantically structured representation it express informative tree fragments by analyzing a parse tree of the messages, and then utilize external knowledge bases to increase their semantic information. Twitter dataset shows that our framework significantly outperforms existing stateof-the-art methods. This research provide insights into Corona virus fear sentiment progression, and outlines associated methods, implication, limitations and opportunities.
doi:10.30534/ijatcse/2020/221952020 fatcat:tvlslsmkubfsnnbov7ox3pmma4