E-MAIL SPAM DETECTION

Abstract

E-mail is the most important form of communication. Used for a wide range of people including individuals and organizations. But these people using this e-mail they find it difficult to use because of spam mail. These spam emails are also called unsolicited bulk mail or junk mail. Spam emails are available randomly sent messages to people by anonymous users. Sites are trying to steal yours personal, electronic and financial information. An increase in spam emails leads to crime of theft of sensitive information, reduced productivity. Spam detection is dirty. The line between spam and non-spam messages is blurred, and the condition changes over time. From various attempts to automate spam detection, machine learning has so far proven to be the most effective and popular method of email providers. While we still see spam emails, a quick look at the trash folder will show how many spam is removed from our inbox daily due to machine learning algorithms. It is estimated that 40% of emails are spam mail. These spam wastes time, storage the space and width of the communication band. There are a few ways to receive spam emails but spam senders make it difficult for you to send users from a random sender address or by adding special characters at the beginning or end of the email. There are several machine learning methods for filtering spam emails including Naïve Bayes classifier, Vector support equipment, Neural Networks, Close Neighbour, Rough Sets and Random Forests. In this project we use the Naïve Bayes classifier to identify spam mail. The vast majority of people depend on what is available email or messages sent by a stranger. Possibly anyone can leave an email or message provide gold the opportunity for spam senders to write a spam message about us different interests. Spam fills in the inbox with a number of funny things mails. Slow down our internet speed. Theft useful information such as our details on our contact list. Identifying these people who post spam and spam content can be a a hot topic for research and strenuous activities. Email Spam is functionality of mass mailings. From the cost of Spam is heavily censored by the recipient, it is a successful post proper advertising. Spam email is a form of commercial advertising economically viable because email can be costly effective sender method. With this proposed model some message may be declared spam or not use Bayes' theorem and Naive Bayes’ Classifier and IP addresses of sender is usually found.