Predicting Nodes for Effectively Spreading Vaccine Awareness: An Influence Maximization Approach

Abstract

Many large pandemics have occurred throughout human history, and pandemic-related crises have had massive detrimental effects on global health, economics, and even national security. Aside from the potentially lethal spread of infections, one key problem is persuading people to get vaccines. For decades, problematic negative vaccine beliefs have been the largest predictor of opposition to vaccination efforts. In this research, we attempted to examine the probability of an individual to be vaccinated or not and offer a strategy for efficiently spreading awareness among people via a well-executed network. Finding a set of seed nodes that maximizes the spread of influence in a social network is known as influence maximization. The seed nodes are employed in viral marketing to maximize profit by leveraging effective word-of-mouth. By finding the most important seeds in the network, we want to use this strategy to raise vaccination awareness in communities. For prediction, we compared a few deep learning models and a random forest classifier. Then we examined the model with the best performance and conducted a basic study on employing influence maximization to raise awareness in distinct communities. The best model was roughly eighty four percent accurate and forecasted us those who are most likely to be vaccinated, which could be utilized as seeds for our Influence Maximization model, to spread vaccination awareness among the unvaccinated, which can also be discovered by our prediction model.