A Comparative Analysis of Logistic Regression and Random Forest Algorithm for Cyber Attacks Classification
Ruuhwan Ruuhwan, Rikza Fauzan Nuradiah, Aso Sudiarjo, Dhema Yunautama, Gunawansyah Gunawansyah, Edi Andriansyah · 2024
This research aims to compare the performance of Logistic Regression and Random Forest algorithms in classifying cyber-attack types. Using a data set consisting of 494,021 data points with 42 attributes, an evaluation was conducted to assess the accuracy, precision, recall, and F1-score of both algorithms. The results of the analysis show that Random Forest is superior to Logistic Regression in classifying cyber-attack types, with the accuracy of the Random Forest algorithm on 80:20 and 70:30 ratios resulting in a value of 99% compared to the accuracy of Logistic Regression on 80:20 ratio resulting in a value of 91% and on 70:30 ratio data can produce an accuracy value of 81%. This research provides a deeper insight into the ability of both algorithms to overcome the challenges of classifying cyber-attack types, which can be valuable for the development of a more effective cyber-attack detection system.