A Comparative Analysis of Machine Learning Techniques for Spam Detection
Syed Ishfaq Manzoor · International Journal of Advanced Trends in Computer Science and Engineering · 2019
Data Science is an emerging multidisciplinary field which employs algorithms, processes, scientific methods to extract information and insights in various forms which is both structures and unstructured much similar to data mining and prediction analysis.Advertisement and bulk emails, also called as spam, makes an estimate of 62% of the Worldwide internet traffic.Since 1978, when first unwanted mail was sent, technology have advanced but still the detection of spams remains a chronophagous and big budget problem in the field of mathematical sciences.The current study evaluates the effectiveness and efficiency of various machine learning techniques which include K-NN, Decision tree, random forest, Naive Bayes and SVM for spam detection.A data set comprising of 962 emails containing both genuine emails and spams has been used in this study.Some deep learning techniques for classification of spams is also suggested for better performance.