Spam image filtering algorithm / Muhammad Hazim Shahimi

Shahimi, Muhammad Hazim (2015) Spam image filtering algorithm / Muhammad Hazim Shahimi. Degree thesis, Universiti Teknologi MARA, Perak.


Email has become the most important medium of transferring message on the internet. The email users are increasing over the years because it is easy to use and low cost. However, this situation has attracted spammer to advertise their product by sending spam messages to anyone who uses email. As a consequence, the number of spam email has increase unexpectedly. The spam has become a serious problem for the email users. They are flood with a lot of spam in their email. This project is done to help the internet user from being flood by the spam especially spam images. The algorithm in this project will help the existing current spam filtering to enhance the method in filtering spam images. This project applied Optical Character Recognition (OCR) and Bayesian probability to filter the spam images. The evaluation task is done by using formula of Precision, Recall, Error Rate, and Accuracy. Based on the result, the testing dataset has achieved 82% of accuracy. It has shown that the proposed algorithm is good in classifying the images. There are some improvements that can be suggested for future works such as use more data, apply the better OCR technique, use another image feature extraction, and use another classifier.


Item Type: Thesis (Degree)
Email / ID Num.
Shahimi, Muhammad Hazim
Email / ID Num.
Thesis advisor
Masrom, Suraya
Subjects: Q Science > QA Mathematics > Mathematical statistics. Probabilities
Q Science > QA Mathematics > Instruments and machines > Electronic Computers. Computer Science
Q Science > QA Mathematics > Instruments and machines > Electronic Computers. Computer Science > Algorithms
Divisions: Universiti Teknologi MARA, Perak > Tapah Campus > Faculty of Computer and Mathematical Sciences
Programme: Bachelor of Computer Science (Hons)
Keywords: Spam image; filtering; algorithm
Date: January 2015
Edit Item
Edit Item


[thumbnail of 57789.pdf] Text

Download (186kB)

Digital Copy

Digital (fulltext) is available at:

Physical Copy

Physical status and holdings:
Item Status:
On Shelf

ID Number




Statistic details