Detection of Malicious E Mails by Improved Artificial Neural Network

E-mail has become one of the today‟s standard means of communication. E-mail data is also growing rapidly and is going to be misused due to its popularity. However, security loopholes in it enable cyber criminals to misuse it by forging its headers or by sending secretly for illegal purposes. In effect, an overwhelming amount of spam is flowing into user‟s mail boxes daily. E-mail spam known as Unsolicited Commercial E-mail (UCE) or junk mail, is the practice of sending

unwanted E-mail messages, frequently with commercial content for a recipient in large quantities. These junk mails can contain various types of messages such as commercial advertising, doubtful product, viruses or quasi legal services. The inverse of "spam" E-mail is called "ham" which needs by recipients. Links on social networks that take you to free downloads, easy income and unsolicited text messages offering loans, low priced products and etc. Spam causes annoyance and wastes user‟s time to regularly check and delete this large number of unwanted messages.

Flooding of mailboxes with spam E-mails waste storage space and overload the server; thus it may lead to losing legitimate E-mails, degrading the server performance, or even make it totally unavailable. Hence, spam consumes network bandwidth and server storage space. In order to overcome this automatic spam filtration system for every individual user is highly essential.