Open AccessA Comparitive Study of E-Mail Spam Detection using Various Machine Learning Techniques
Author(s) -
Simarjeet Kaur,
Meenakshi Bansal,
Ashok Kumar Bathla
Publication year - 2021
Publication title -
aijr proceedings
Language(s) - English
Resource type - Conference proceedings
ISSN - 2582-3922
DOI - 10.21467/proceedings.114.56
Subject(s) - computer science , naive bayes classifier , support vector machine , machine learning , random forest , artificial intelligence , statistical classification , feature extraction , electronic mail , set (abstract data type) , bag of words model , data mining , world wide web , programming language
Due to the rise in the use of messaging and mailing services, spam detection tasks are of much greater importance than before. In such a set of communications, efficient classification is a comparatively onerous job. For an addressee or any email that the user does not want to have in his inbox, spam can be defined as redundant or trash email. After pre-processing and feature extraction, various machine learning algorithms were applied to a Spam base dataset from the UCI Machine Learning repository in order to classify incoming emails into two categories: spam and non-spam. The outcomes of various algorithms have been compared. This paper used random forest, naive bayes, support vector machine (SVM), logistic regression, and the k nearest (KNN) machine learning algorithm to successfully classify email spam messages. The main goal of this study is to improve the prediction accuracy of spam email filters.