Detection of Malware Attacks Using Naive Bayes Algorithm Comparing Logistic Regression Algorithm to have Improved Accuracy Rate

Abstract

Aim:  To enhance the accuracy in Detection of Malware in Detection of Malware attacks Using Naive Bayes Algorithm comparing Logistic Regression Algorithm to have improved accuracy rate. Materials and Methods: This study contains 2 groups i.e novel Naive Bayes Algorithm (NB) comparing Logistic Regression Algorithm (LR).  Each group consists of a sample size of 30. Their accuracies are compared with each other using different sample sizes also. The G-Power in the test set will be at 80%. Results: Data is trained in the given model so that Machine learning can function effectively. The  Logistic Regression Algorithm is 50% more accurate than the Naive Bayes Algorithm of 62.8% in classifying the malware Detection.The outcomes have been acquired with a stage of importance fee of p=0.053, with a pretest power value of 80% using SPSS tools. Conclusion: Through this, Prediction is done for The Naive Bayes model is significantly better than the Logistic Regression in identifying Novel Malware Attacks Analysis. Naive Bayes can be also considered as a better option for the classification of Novel Malware Attacks Analysis.