Comparative Analysis of GUI based Spam Message Classifier using Machine Learning Approach

Abstract

Generally, Spam messages are sent randomly or particularly to addresses by lazy advertisers and phishing criminals who wish to lead people to malicious and phishing sites. Spam detection is a significant application of Machine Learning on the internet today. Like a lot of other applications, machine learning models can be trained to distinguish between non-spam (ham) emails and spam. So, the aim is to examine and survey machine learning algorithms to identify one or multiple as best or better techniques to use in content-based spam filtering. Current spam techniques could be paired to increase effectiveness and to investigate ma-chine learning-based techniques for spam prediction results with the best accuracy. The analysis by supervised machine learning to capture information like variable identification, univariate, bivariate and multivariate analysis, data validation, cleaning, preparing, visualization is done on the entire dataset. This analysis will be aundogmatic guide to model parameters’ sensitivity analysis about performance in the prediction of spam mails by utilizing the corresponding accuracy calculation. Additionally, the comparison of the performance of various machine learning algorithms from the given dataset with an evaluation of classification re-port, sensitivity, specificity, confusion matrix, and different other score metrics is performed to create a better picture of this evaluation. The result will show that the efficacy of proposed machine learning algorithm techniques can be compared with the best precision with Accuracy, Precision, Recall, and F1 Score.

Keywords:Dataset, Python, Machine Learning, Spam, Spam Classification, Accuracy Result, Comparison, GUI.