Hybrid Spam Filtration Method using Machine Learning Techniques
S. Jancy Sickory Daisy1, A. Rijuvana Begum2
1S. Jancy Sickory Daisy, Department of Computer Science Engineering, PRIST University, Thanjavur-613403.
2A. Rijuvana Begum, Department of Engineering and Communication Engineering, PRIST University, Thanjavur-613403
Manuscript received on 20 June 2019 | Revised Manuscript received on 05 July 2019 | Manuscript published on 30 July 2019 | PP: 1818-1821 | Volume-8 Issue-9, July 2019 | Retrieval Number: I8974078919/19©BEIESP | DOI: 10.35940/ijitee.I8974.078919
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Electronic mail (e-mail) is one of the most prevalent approaches for online communication and transferring data through web because of its quick and easy distribution of data, low distribution cost and permanency. Despite these benefits there are certain weaknesses of e-mail. Among these, spam also known as junk e-mail tops. Spam is set of unwanted or inappropriate messages sent over the internet to a massive amount of users for the purpose of marketing, phishing, disseminating malware, etc.With the internet becoming the dominant platform anti-spam solutions are of great use today. This paper illustrates an efficient hybrid spam filtration method using Naïve Bayes algorithm and Markov Random Field technique, which detects and filters spam messages. The proposed method is evaluated based on its accurateness, meticulousness and time consumption. The results confirm that the proposed hybrid method achieves high percentage of true positive rate in finding e-mail spam messages.
Keywords: E-mail Spam, Naïve Bayes Algorithm, Markov Random Field.
Scope of the Article: Machine Learning