Author :
Ketari, Lamia Mohammed ; Chandra, Munesh ; Khanum, Mohammadi Akheela
Author_Institution :
Dept. of IT, King Saud Univ., Riyadh, Saudi Arabia
Abstract :
Nowadays, Internet is becoming an incontestable communication mean of choice, through emails, for both professional and personal correspondences. However, emails are not only, efficient, cheap, and rapid mean of communication, but also are becoming favourite lucrative business spammers. According to many survey reports, the amount of unsolicited e-mails, known as spam e-mails, is becoming a great and a serious problem because of the huge losses they yield to the organizations, ranging from extensive bandwidth consumption, mail server processing load, to user´s productivity due to time spent on detection and dealing with spam mails. Initially, spam e-mails contained only textual messages and were easily detected by the text-based spam filters. To avoid detection, spammers have come up with a new approach to send their spam. It consists in including their advertisements as part of an embedded image file attachment (.gif, .jpg, .png, etc.) rather than the body of the e-mails hence defeating text-based spam filtering techniques. This paper investigates the major image spam filtering techniques in current use. While studying these techniques, both success rates and possible problems are explored. The relevant work shows that spammers are becoming more sophisticated in their approach to adapt to all challenges, hence defeating the conventional spam filtering techniques. Consequently, image spam detection techniques should be scalable and adaptable to meet the future spammers´ tactics.
Keywords :
Internet; business data processing; document image processing; information filtering; text analysis; unsolicited e-mail; Internet; bandwidth consumption; business spammer; embedded image file attachment; image spam detection; image spam filtering; mail server processing load; organization; spam e-mail; text-based spam filtering; textual message; unsolicited e-mail; user productivity; Accuracy; Feature extraction; Filtering; Image color analysis; Support vector machines; Unsolicited electronic mail; classification; detection techniques; filtering; image spam; spam;