Detection of Email Spam using an Ensemble based Boosting Technique
Uma Bhardwaj1, Priti Sharma2
1Uma Bhardwaj, Department of Computer Science & Applications, Maharshi Dayanand University, Rohtak, India.
2Priti Sharma, Department of Computer Science & Applications, Maharshi Dayanand University, Rohtak, India.
Manuscript received on 28 August 2019. | Revised Manuscript received on 14 September 2019. | Manuscript published on 30 September 2019. | PP: 403-408 | Volume-8 Issue-11, September 2019. | Retrieval Number: K13650981119/2019©BEIESP | DOI: 10.35940/ijitee.K1365.0981119
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: Email is amongst the advanced socializing communication source that is mostly adapted by business and commercial users. Although there are various assets of email as it is quick, cheap, and efficient communication resource, the service is misused by various users for the personal or professional purposes by spreading the useless and extra emails which are termed as email spam. There is the existing research of authors who have used machine learning methods for the detection of email spam and achieved effective precision values, but the individual methods mostly noted with some shortcomings which leads to lack in the performance. In this research work, ensemble based boosting technique on machine learning classifiers of Multinomial Naïve Bayes and J48 classifiers is proposed for the detection of email spam. The system detects email spam by adding the strong features of one classifier into another with the help of Adaboost algorithm. The results of the proposed ensemble technique are evaluated with accuracy and sensitivity parameters. The evaluated results indicate the effectiveness of proposed concept to detect the email spam in comparison with individual methods and other considered existing concepts.
Keywords: Adaboost, Boosting, Decision tree J48 algorithm, Email spam, Multinomial Naïve Bayes, Spam filtration, Text analysis.
Scope of the Article: Classification