Efficient Prediction In Improving Accuracy For U2R Attack In Wireless Networks Using Novel CNN Alexnet In Comparison With Support Vector Machine Algorithm
V Chandra Sekar, V. Nagaraju · 2024
The major goal of this experimental effort is to use DL (Deep Learning) models to improve the accuracy value of User-to-Root (U2R) attack forecasting using the intrusion dataset. The Github database system served as the source of the research data bank used in this study. A sample size of 22543 with 30 iterations was used to forecast a User-to-Root (U2R) attack (15 from Group 1 and 15 from Group 2) which was obtained with a G-power of 0.8, alpha and beta values of 0.05 and 0.2, and a confidence range of 95%. Novel Convolutional Neural Network Alexnet (CNN Alexnet) and Support Vector Machine are used to predict the User-to-Root (U2R) attack of the infiltration with the same number of data samples (N=15), with Novel Convolutional Neural Network Alexnet (CNN Alexnet) having a higher accuracy rate. Performance evaluation of the proposed U2R attack prediction system of the Wireless Network system. The accuracy level of the Novel Convolutional Neural Network Alexnet (CNN Alexnet) classifier is confirmed at 92.4060%, and the accuracy level of the SVM is confirmed at 87.4387%. These two can be distinguished clearly from one another. The value of significance is found to be p=0.525 (p>0.05) after analyzing the results from the independent samples test and hence it is statistically not significant. The built Novel Convolutional Neural Network Alexnet (CNN Alexnet) model performs better in User-to-Root (U2R) attack prediction from intrusion data banks than the SVM classifier, according to this developed research design outcome.