Email Security Classification of Imbalanced Data Using Naive Bayes Classifier
Alphy Abraham · International Journal of Wireless Communications and Network Technologies · 2019
Email is a widely accepted communication method even in large cooperates.However, it is not a safe platform for the companies due to the risk of information leakage, spamming or privacy threat.Corporate-based email systems have different standard etiquette's.It should ensure that the personal information of a company is not being leaked.It is a very hard task for the cyber team to manually check and verify the outbound emails from a company that the private information or important documents of the company is kept safe.So, in the present work we introduce a system to scrutinize the number of outbound emails that the manager has to check by classifying them according to their security levels.This inturn curtails the effort for the cyber security team within the company.The email dataset of all the employees within the company is first obtained in the form of a csv file and is then fed into the classifier to train them and classify new mails according to their security levels.The emails are pre-processed first before classifying them in real time.The high security mails are then send to the manager for verification.And the mails are either forwarded or discarded.