A DNN Architecture Generation Method for DDoS Detection via Genetic Alogrithm

Abstract

Nowdays, DNNs (Deep Neural Networks) are widely used in the field of DDoS attack detection. However, designing a good DNN architecture relies on the designer’s experience and requires considerable work. In this paper, a GA (genetic algorithm) is used to automatically generate the DNN architecture for DDoS detection to minimize human intervention in the design process. Furthermore, given the complexity of contemporary networks and the diversity of DDoS attacks, the objective of this paper is to generate a DNN model that boasts superior performance, real-time capability, and generalization ability to tackle intricate network scenarios. This paper presents a fitness function that guarantees the best model generated possesses a specific level of real-time capability. Additionally, the proposed method employs multiple datasets to joint models generated, thereby enhancing the model’s generalization performance. This paper conducts several experiments to validate the viability of the proposed method. Firstly, the best model generated with one dataset is compared with existing DNN models on the CICDDoS2019 dataset. The experimental results indicate that the model generated with one dataset has higher precision and F1-score than the existing DNN models. Secondly, model generation experiments are conducted on the CICIDS2017 and CICIDS2018 datasets, and the best model generated still performs well. Finally, this paper conducts comparative experiments on multiple datasets using the best model generated with six datasets and the best model generated by existing methods. The experimental results demonstrate that the best model generated with six datasets has better generalization ability and real-time capability.

Assessment of Landslide Susceptibility in the Moxi Tableland of China by Using a Combination of Deep-Learning and Factor-Refinement Methods

“…It employs a loss function to quantify the discrepancy between the network's predicted output and the actual target. Subsequently, the network undergoes iterative training through the application of the backpropagation algorithm and an optimization technique, iterating extensively to fine-tune the network weights and thereby minimize the loss function [21].…”

A Comprehensive Survey of Distributed Denial of Service Detection and Mitigation Technologies in Software-Defined Network

“…Traditional machine learning SVM [31][32][33][34][35] Decision Tree [36][37][38] KNN [38][39][40][41] Naive Bayes [38,[42][43][44] Random Forest [36][37][38] Deep learning SOM [41,45,46] ANN [47][48][49] LSTM [48][49][50] DNN [51][52][53] RNN [50,53] The SVM algorithm is a binary classification model utilized for distinguishing between normal and abnormal data in the context of DDoS attack detection based on traffic characteristics. Based on the traffic characteristics observed in the SDN network environment, the SVM detection algorithm is employed to gather input feature vectors in order to develop an algorithm for detecting malicious behavior within the network.…”

Pendekatan Metode Ensemble Learning untuk Deteksi Serangan DDoS menggunakan Soft Voting Classifier

“…Pada penelitian yang menggunakan machine learning untuk deteksi DDoS, digunakan sejumlah data yang diekstraksi oleh para ahli untuk melatih dan menguji model guna meningkatkan kinerja. Sedangkan pada penelitian yang menggunakan deep learning untuk penelitian DDoS biasanya memungkinkan model untuk memilih fitur terbaik untuk melatih dan menguji model pada data tersebut [7].…”

“…Pemilihan fitur atau seleksi fitur merupakan langkah penting dalam pembelajaran mesin yang melibatkan pemilihan sekelompok fitur yang penting dari sekumpulan fitur yang lebih besar untuk meningkatkan efisiensi model [7]. Proses ini sangat mempengaruhi kinerja model dengan mengurangi kompleksitas dimensi, meningkatkan kemampuan generalisasi, mempercepat proses pembelajaran, dan meningkatkan kemampuan untuk menjelaskan model [7].…”