Security concerns and economic losses caused by network attacks have inspired extensive network security research with active study fields that include data collection, analysis, and visualization. Visualization can help analysts to efficiently detect unusual behavior patterns in a vast amount of data, which should result in a prompt response to a potential security threat. However, the majority of the research papers suggest custom visualization solutions for the proposed analysis techniques rather than using available and well-adopted data pipelines that can further support the analysis. In this paper, we propose a system architecture for clustering, visualization, and computer-assisted network security analysis based on an open-source Elastic Stack. We extended the data pipeline in order to enable data clustering prior to visualization and employed visual mapping techniques to filter data and successfully spot the network assaults. A case study that demonstrates the effectiveness of the proposed solution is given in the example of a port scan attack.
System and network security have become one of the main considerations for the system or network design, implementation and maintenance. The increased volume of traffic and the ever-changing form of internet communications has led to a steep rise in network attacks, hence the need to protect both sides of the communication is essential. Over the last couple of years, an upsurge in the use of machine learning and deep learning techniques in detecting network attacks can be noticed. Alongside with individual machine and deep learning techniques, various ensemble techniques have been implemented in order to combine multiple models working on the same problem. However, ensemble techniques mostly focus on machine learning models, disregarding the fact that deep learning techniques have been proven to be efficient when handling large amounts of data. In this paper, a network intrusion detection system is implemented using a weighted voting ensemble deep learning model. Fifteen various Deep Neural Network models have been trained and evaluated on the imbalanced multiclass CICIDS-2017 dataset. Weighted voting system has been implemented for combining decisions of various models in two ways. The evaluation results for multiple combinations are given. The proposed system has the ability to use heterogeneous models for the decision.
Given the ever-increasing volume of traffic and the developing nature of internet communications, security has emerged as one of the primary considerations in the design and implementation of any system or network. The frequency of network attacks is increasing, so the need for protecting both the users’ and systems’ data is crucial. In recent years, there has been an increase in use of machine learning and deep learning techniques as an aid for detecting network attacks. Various deep learning models have been in focus of researchers for the past couple of years, having in mind their capabilities for processing large amounts of data. However, when it comes to handling multi-class imbalanced data, proper detection of multiple attacks is still a challenge. In this paper, a multiclass network intrusion detection system is implemented using a convolutional neural network design. The CICIDS-2017 dataset, a highly imbalanced dataset with 15 distinct traffic classes, is used to train and test the model. Two feature selection techniques are used, and depending on the input dataset dimensions and the hyperparameter variations, 24 different CNN model variations are trained. In order to evaluate the models objectively, macro average metric is used. A comparison is made with a DNN model, in order to show the CNN model capabilities. It is shown that the proposed model outperforms DNN and detects multiple classes that are represented by a limited number of inputs in the dataset with high precision.