A Potent Technique for Identifying Fake Accounts on Social Platforms

Abstract

In this generation, online social media networks are rapidly growing in popularity and becoming more and more integrated into people's daily lives. These networks are used by users to exchange movies, read news articles, market products, and more. It has been simpler to add new friends and stay in touch with them and their updates. These online social networks have been the subject of research to see how they affect people. A significant amount of a user's data may attract attackers as these networks continue to develop, and these attackers may subsequently exchange incorrect information and disseminate dangerous falsehoods. Some fraudulent accounts are used to spread false information and further political agendas, for example. Finding a fraudulent account is important. Furthermore, these social networking platforms are increasingly being used by attackers to disseminate a vast amount of fake information. As a result, based on the categorization algorithms, researchers have started to investigate efficient strategies for spotting these sorts of actions and bogus accounts. In this study, various machine learning algorithms are investigated to successfully identify a phony account. To address this issue, several machine learning algorithms are utilized in conjunction with pre-processing methods to identify bogus accounts. The identification of bogus accounts uses the classification abilities of the algorithms Nave Bayes, Artificial Neural Network, Bagged Decision Tree, Radial Basis Function (RBF), Support Vector Machines, and Random Tree. The best features are used to compare the proposed model to other benchmark techniques on the dataset. The suggested Artificial Neural Network strategy outperforms the prior employed strategies to identify phony user accounts on major online social platforms, with a precision of 98.90%, when machine learning techniques are also compared.