Abstract:
The COVID-19 pandemic has impacted the world as a whole in ways which were unimaginable before. From the economical and medical impacts to geopolitical views and influences, COVID-19 has changed the world as we see it. Since the introduction of different vaccines to prevent COVID-19, people’s opinions have been divided regarding it. The social media platform, Twitter, provides a noteworthy platform for voicing opinions in support of and against the vaccines which results in long debates and discussion and often spreading of misinformation. In this paper, a dataset has been manually collected from twitter using the Twitter API and tweets were manually annotated into three distinct categories – provac, antivac and other. Six machine learning algorithms were used to train and test on the annotated data and the best classifier for this case was identified. Using the best classifier, the whole dataset was automatically annotated and stance towards the COVID-19 vaccine was analyzed. Further analysis was done to identify changes in trends of people’s opinions over time. The results indicate that, with proper implementation of the ML algorithms, it is possible to identify and predict people’s stances towards the COVID-19 vaccine and similar approach can be used in analyzing stance towards other vaccines and treatments of various diseases.
Description:
Supervised by
Mr. Safayat Bin Hakim,
Assistant Professor,
Department of Electrical and Electronic Engineering(EEE),
Islamic University of Technology(IUT),
Board Bazar, Gazipur-1704, Bangladesh.
This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Electrical and Electronic Engineering, 2022.
