Machine Learning-based spam detection using Naïve Bayes Classifier in comparison with Logistic Regression for improving accuracy

Abstract

Aim: The aim of this research is to detect spam using machine learning with the Novel Naive Bayes Classifier (NB) and the Logistic Regression (LR). Material and Methods: For analyzing the spam, it needs two groups which consist of 40 samples. The two groups are group 1 which consists of Novel Naive Bayes Classifier (NB) with a sample size of 20 and group 2 which consists of Logistic Regression (LR) with a sample size of 20 and G-power (value = 0.8). Results: Novel Naive Bayes Classifier has an accuracy of 98.05% which is comparatively more than the Logistic Regression with an accuracy of 94.7%. The accuracy has a 2-tailed significant value of 0.012 (p<0.05) which is found in the Independent Sample T-Test analysis. Conclusion: The performance of the Novel Naive Bayes Classifier is more than the performance of Logistic Regression in terms of accuracy.