Malevolent Application of Bots-Detecting and Forecasting the Prospect of Bots Trigger-ing Misinformation Spread Using the Machine Learning Approach

Abstract

Despite the fact that social media platforms like Facebook and Twitter, offer implausible opportunities to their users, social media remains a major part of our everyday life. Because they are immensely popular among a wide range of types of users, social media platforms, such as Facebook, are increasingly used by automated accounts, also acknowledged as bots. The key work of these bots is disseminating fake news, promoting specific ideas and products, manipulating the stock market, and even distributing sexually explicit materials. On the other hand, machine learning models have successfully detected fake accounts created by bots or computers in many instances. Machine Learning algorithms such as SVM, Naïve Bayes, Bayes Network, Neural Network, and other classifiers that were used for Bots detection were compared for precision, readability, accuracy. An in-depth review of systematic literature discusses the supervised machine learning classifiers that make use of labeled data to train. From the perspective of the context, we examine the previous literature examining the role of bot-generated accounts that spreads such misinformation also known as fake news, similarities to the COVID-19 pandemic. By examining known bots’ prediction through machine learning approaches, we conclude by stating the prevalence of bots in misinformation spread. The majority of Binary classification models are used to detect bots and the major limitation is the model’s capacity to figure out the correspondence between the reported news and the actual news, so we suggest certain ideas to sort this out.