USE OF NAIVE BAYES CLASSIFIER FOR SPAM FILTERING

Use of electronic mail (e-mail); Since the spread of internet usage has become simple and easily accessible, it has grown in the last forty years and has become one of the most widely used communication tools today. Increased use of email has brought with it a number of problems. One of the major problems that arise is unwanted, advertising-containing electronic mail. These disturb the users of the email and also cause unnecessary resource waste. Advertisement-based e-mails are used to promote law or illegal products, and millions of unsolicited e-mails that are sent from different sources for different purposes fill the mail boxes of Internet users. Commercial e-mails, which become increasingly a problem, are engaging both internet traffic and mail servers. A lot of work has been done on the filtering of unwanted emails. In this study, a spam filter was designed using Naive Bayesian Classification. Instead of focusing on increasing spam sensitivity, it is intended to protect all non-spam e-mail as the first priority.

Anahtar Kelimeler:

Spam, ham, NaiveBayesClassifier, e-mail

USE OF NAIVE BAYES CLASSIFIER FOR SPAM FILTERING

Keywords:

Spam, ham, NaiveBayesClassifier, e-mail,

PDF

___

1. Ketari L.M., Chandraand L.M., Khanum M.A. ‗A Study of image Spam Filtering Techniques.‘4th IEEE Internet. Conf. Computatio nalIntelli genceand Communication Networks,2012.
2. Deshmukh S.S., Chandre P.R. ‗Survey on: Naïve Bayesianand AOCR Based Image and TextSpam Mail Filtering System.‘ International Journal of Emerging Technology and Advanced Enginerring.
3.Shcherbakova T., Vergelis M., Demidova N. ‗SpamandPishing in the First Quarter of 2015‘. https://securelist.com/spam-and-phishing-in-the-first-quarter-of-2015/69932/ [May.13,2015]
4. Mark B. ‗8 emailstatisticstouse at parties‘. Sponsoredby: 2013 Email Marketing Bechmark Report. http://www.email-marketing-reports.com/iland/2009/08/8-email-statistics-to-use-parties [Jul.17,2015]
5. Sahami M., Dumais S., Heckermanand D. Horvits E. 1998. ‗A. Bayesian Approachto Filtering Junk E-mail‘. In. Froc, of the AAAI‘98 Workshop on Learning for Text Categorization, pp. 1048-1054
6. Livington B. ‗Paul Grahampro videsstunningans werto spam e-mails‘. http://www.infoworld.com/article/2674702/technology-business/paul-graham-provides-stunning-answer-to-spam-e-mails, 2002.
7.Paul Graham, Better Bayesiy anfiltering. http://www.paulgraham.com/better-bayesian-filtering, 2003.
8. Fitre L., Lemnaru C., Potolea R. ‗Spam Detection Filter Using KNN Algorithmand Resampling‘ in Intelligent Compute Communication and Processing (ICCP),2010 IEEE International Conferenc., 2010 pp. 27-33.
9. Kyriakopoulouand A., Kalamboukis T. ‗Text Classification Using Clustering‘, in ECML-PKDD Discovery Challange Workshop Proceedings,2006.
10.Dylan Mors & Dermot Harnett. "State of Spam, a Monthly Report - Report #33" (PDF) 2009.
11.Process Software, Introduction to Bayesian Filtering
12.https://en.wikipedia.org/wiki/Naive_Bayes_spam_filtering
13.Brian Burton (2003). "SpamProbe - Bayesian Spam Filtering Tweaks".
14.Jonathan A. Zdziarski (2004). "Bayesian Noise Reduction: Contextual Symmetry Logic Utilizing Pattern Consistency Analysis".
15.Machine Learning with R. Packt Publishing. Bitmingham-Mumbai.2013,pp. 89 - 117