Applications of sentiment analysis and machine learning techniques in disease outbreak prediction – A review

Abstract

Prediction of the facts about any topic always sounds fascinating. It becomes more important and useful when it can predict the facts about healthcare information. In computer science, machine learning techniques/models can be used to predict important health care facts, further the study of the sentiments related to health care information can also play a significant role in the generation of important information. There are many international and national organizations like United States Centers for Disease Control and Prevention (CDC), World Health Organization (WHO), etc. that publish important healthcare data. These data sets can be used to predict important facts about the disease. Further, the datasets related to sentiments of the people like tweets on Twitter, posts on Facebook, and blogs are useful to study the sentiments of the people on a particular topic. Sentiments related to health care information, diseases, and epidemics are very useful. The study of these sentiments can help make better prediction systems and to generate fruitful health care facts. Further, the real datasets related to infectious diseases like COVID-19 can be used for regression analysis for making predictions regarding the upcoming cases. In this paper, we have presented a survey on the application of different machine learning techniques, that are used in the classification and regression analysis related to health care predictions and the study of sentiments. We have presented a review of major research studies related to healthcare between the years 2010 to 2020.