Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

K-Nearest Neighbor Classification Of E-Mail Messages For Spam Detection


Affiliations
1 Department of Computer Science, Bharathiar University, India
     

   Subscribe/Renew Journal


The increase in unwanted spam email volumes created a clear need for more effective and robust anti-spam filters. Recent machine learning methods are employed to detect and process spam mails successfully. In this paper, we present a density based clustering for email classification problem using kNN algorithm. Initially, the relevant features for filtering the spam messages are extracted from the study and it acts as an antispam filter. It thereby generates the successful corpus list for detection of spam emails. The experiments are conducted on various email datasets and the results show that the proposed kNN density based clustering offers improved performance than the other methods. As shown by the test results, our methodology showed stronger prediction capabilities and better classifications based on in-depth learning techniques.

Keywords

Email, Spam Classification, Feature Extraction, Corpus.
Subscription Login to verify subscription
User
Notifications
Font Size

Abstract Views: 2

PDF Views: 0




  • K-Nearest Neighbor Classification Of E-Mail Messages For Spam Detection

Abstract Views: 2  |  PDF Views: 0

Authors

U. Murugavel
Department of Computer Science, Bharathiar University, India
R. Santhi
Department of Computer Science, Bharathiar University, India

Abstract


The increase in unwanted spam email volumes created a clear need for more effective and robust anti-spam filters. Recent machine learning methods are employed to detect and process spam mails successfully. In this paper, we present a density based clustering for email classification problem using kNN algorithm. Initially, the relevant features for filtering the spam messages are extracted from the study and it acts as an antispam filter. It thereby generates the successful corpus list for detection of spam emails. The experiments are conducted on various email datasets and the results show that the proposed kNN density based clustering offers improved performance than the other methods. As shown by the test results, our methodology showed stronger prediction capabilities and better classifications based on in-depth learning techniques.

Keywords


Email, Spam Classification, Feature Extraction, Corpus.