An Improved Spam Detection Method with Weighted Support Vector Machine

Email is the most admired method of exchanging messages using the Internet. One of the intimidations to email users is to detect the spam they receive. This can be addressed using different detection and filtering techniques. Machine learning algorithms, especially Support Vector Machine (SVM), can...

Full description

Saved in:
Bibliographic Details
Published in2018 International Conference on Data Science and Engineering (ICDSE) pp. 1 - 5
Main Authors Vishagini, V, Rajan, Archana K
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.08.2018
Subjects
Feature extraction
Kernel
KFCM
KPCM
Prediction algorithms
Spam detection
Support vector machines
SVM
Training
Unsolicited electronic mail
Weighted SVM
Online AccessGet full text
DOI10.1109/ICDSE.2018.8527737

Cover

More Information
Summary:Email is the most admired method of exchanging messages using the Internet. One of the intimidations to email users is to detect the spam they receive. This can be addressed using different detection and filtering techniques. Machine learning algorithms, especially Support Vector Machine (SVM), can play vital role in spam detection. We propose the use of weighted SVM for spam filtering using weight variables obtained by KFCM algorithm. The weight variables reflect the importance of different classes. The misclassification of emails is reduced by the growth of weight value. We evaluate the impact of spam detection using SVM, WSVM with KPCM and WSVM with KFCM.UCI Repository SMS Spam base dataset is used for our experimentation.
DOI:10.1109/ICDSE.2018.8527737