Probabilistic Inference and Trustworthiness Evaluation of Associative Links toward Malicious Attack Detection for Online Recommendations

Abstract

Today, spamming mails is one of the biggest issues faced by everyone in the world of the Internet. In such a world, email is mostly shared by everyone to share the information and files because of their easy way of communication and for their low cost. But such emails are mostly affecting the professionals as well as individuals by the way of sending spam emails. Every day, the rate of spam emails and spam messages is increasing. Such spam emails are mostly sent by people to earn income or for any advertisement for their benefit. This increasing amount of spam mail causes traffic congestion and waste of time for those who are receiving that spam mail. The real cost of spam emails is very much higher than one can imagine. Sometimes, the spam emails also have some links which have malware. And also, some people will get irritated once they see their inbox which is having more spam mails. Sometimes, the users easily get trapped into financial fraud actions, by seeing the spam mails such as job alert mails and commercial mails and offer emails. It may also cause the person to have some mental stress. To reduce all these risks, the system has proposed a machine learning model which will detect spam mail and non-spam emails, and also this system will optimize the data by removing the unwanted mails which contain the advertisement mails and also some useless emails and also some fraud mails. This proposed system will detect the spam mails and ham emails with the dataset consisting of spam mails and after identifying spam mails this system will remove that spam emails and this proposed system will calculate the amount of storage before and after the removal of spam mails.