Unlocking the Potential of Deep Neural Networks with Tuned Particle Swarm Optimization for Firewall Log Data Analysis: Empowering Network Security Insights

In today’s world, it is vital and in trend to evaluate the firewall of computer networks and control internet traffic based on the research findings. A firewall is an excellent tool that ensures traffic control during the machines communicating on a network. It uses methods defining the traffic according to a set of precise standards and thus helps prevent cyber-attacks which may be costly for an organization. In this paper, we suggested an intelligent method for the classification of incoming and outgoing firewall travel packets. The $\log$ files in this research are classified using Deep Neural Network (DNN) classifiers. This paper introduces Tuned Particle Swarm Optimization (TPSO), an FST for reducing the number of variables in log file datasets. The TPSO proposed model is based on PSO, with the Naive Bayes (NB) classifier serving as the objective function. The approach involves using DNN as a classification model and Tuned PSO as a feature selection technique to enhance the classification performance. The TPSO algorithm is employed to optimize the selection of relevant features from the log-file dataset, aiming to improve the accuracy and efficiency of the classification process. By integrating DNN and TPSO, the study aims to leverage the power of deep learning for capturing complex patterns in the log-file dataset, while also utilizing the optimization capabilities of TPSO to select the most informative features for classification. The combined approach seeks to improve the accuracy and effectiveness of the classification task on the logfile dataset, potentially leading to enhanced insights and decision-making based on the classified log data.