Detecting Spam Email Using Machine Learning
Main Article Content
In everyday life, email became the most affordable also simple method of communication for both official works and business applicants because of simple convenience of web access. The spam mails are increasing with the growth in internet users and Gmail end user. People misuse by sending unwanted emails for commercial purposes and fraudulent purposes. Emails are not the only way to send the spam messages, they can also be found in SMS, forums, social media etc. Some spam contains attachments that if it is opened the computer can be infected with viruses or malware. The data classification method is used in the email filtering. When it comes to data classification, choosing the best-performing classifier is a critical step. Many researches provided many techniques to detect these spam emails and improve the accuracy by using machine learning (ML) algorithms. Both naive bayes (NB) classifier and binomial logistic regression had the option to detect the spam mails as naive bayes can be used to classify large data whereas logistic regression is a statistical method respectively. These algorithms are performed on a ling-spam dataset taken from kaggle website. In the proposed system various datasets are performed on the dataset. The result of the proposed model will be compared with the base models to conclude whether the implemented models have improved the performance and evaluation.
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.