Abstract:
As we all are living in pandemic covid19, which is affecting every perspective of daily life. Every
class is badly suffering from this disease. It has been a subject of discussion since 2019 due to the
increased prevalence of social media and its extensive use and it has been a source of tension, fear,
and disappointment for people all over the world. In this research, we have taken the covid19
tweets data of ten different regions including United States of America, Pakistan, India, China,
South Africa, Philippines, United Kingdom, Switzerland, Philippines, and Ireland with timespan
of 25th July of 2020 to 29th august of 2020. Using the common word embedding technique count vectorizer, we experimented with different classifiers on data to train deep neural networks to
improve the accuracy rate for predicting emotions. We assigned sentiments with highest accuracy
rate. After mining the opinions of these regions about covid19, we have collected the PCR results
of these regions. we have compared the percentage of opinion in form of positive or negative
responses with percentage of per million PCR covid results of these regions. After this region and
time wise twitter-based PCR analysis, we came to know that how much these regions are serious
for pandemic. Also, we figured out a real time international measure to detect these region’s
seriousness for any future pandemic. This research can help the administrations of different regions
for taking wise and suitable steps for controlling spread of any outgoing and upcoming pandemic.
