期刊名称:International Journal of Computer Trends and Technology
电子版ISSN:2231-2803
出版年度:2021
卷号:69
期号:9
页码:12-16
DOI:10.14445/22312803/IJCTT-V69I9P103
语种:English
出版社:Seventh Sense Research Group
摘要:In recent years, the increasing use of Electronic mail for fast and cheap personal, official, academic communication, and electronic commerce has led to the emergence and further widespread of problems caused by unsolicited and unwanted bulk e-mail messages. In this study, the objective is to enhance the classification of incoming e-mails-using the Naïve Bayes classifier-into unwanted and ham (legitimate) based on features in both the Subject text of the email and the Email body. The system segments the input email body into tokens and analyses its structure. The dataset is cleaned, and the total number of unique words are counted and extracted, and then compared with already learned unwanted words in the database. If email is classified as ‘Unwanted with very high degree’ or ‘Unwanted with high degree’, users are notified and advised to block unwanted emails. Some emails were classified as Ham. This means that users can view such messages as legitimate messages.