云安全：集成多滤波器特征选择方法对抗DDoS攻击

需积分: 10 50 浏览量更新于2024-07-16 收藏 458KB PDF 举报

"Ensemble-based Multi-Filter Feature Selection Method for DDoS Detection in Cloud Computing" 在当前云计算广泛应用的时代，网络安全问题日益突出，尤其是分布式拒绝服务（DDoS）攻击，这种攻击通过消耗云服务的带宽、资源和服务，使得云服务提供商和用户无法正常访问。面对海量的数据流量，数据挖掘和机器学习的分类算法被用于识别正常流量与异常流量，以防御DDoS攻击。特征选择作为预处理阶段，可以在监督学习中筛选出原始数据集中的关键特征，提高分类准确性和降低计算复杂性。本研究提出了一种基于集成的多过滤器特征选择方法，该方法结合了四种过滤方法的输出，以实现最优的特征选择。实验在入侵检测基准数据集NSL-KDD上进行，并采用决策树分类器对所提方法进行了详尽的评估。实验结果显示，该方法成功地将特征数量从41个减少到13个，同时保持了高检测率和分类准确性，与其它分类技术相比表现出色。论文深入探讨了在云环境下的DDoS检测问题，强调了特征选择的重要性。所提出的集成方法结合了多个过滤器策略，例如卡方检验、信息增益、互信息等，这些过滤器分别从不同角度评估特征的重要性。通过整合这些方法的结果，可以更全面地评估特征对分类任务的贡献，从而减少冗余特征并提高模型性能。此外，论文还讨论了在云环境中应用此方法可能面临的挑战，如大数据量处理、实时性需求以及模型的可扩展性。在NSL-KDD数据集上的实验表明，特征选择不仅能够提高分类效率，还能减少训练和预测的时间，这对于处理云环境中的大规模数据至关重要。同时，通过对比分析，证明了该方法相比于单一的特征选择方法，其性能有显著提升，这对于提高DDoS攻击检测系统的有效性具有重要意义。本文提出的集成式多过滤器特征选择方法为云环境中的DDoS防御提供了新的思路，通过优化特征集合，提高了机器学习模型的分类效能，有助于构建更加高效且准确的DDoS检测系统。这种方法对于未来云安全领域的发展，特别是在应对复杂网络攻击方面，具有很大的研究价值和实际应用潜力。

methods have been proposed to solve this and discard redundant, irrelevant and noisy

features. In this section, we review works that have been proposed in the literature.

Wang and Gombault [10] propose a filter selection method using IG and Chi-squared to

extract 9 most important features from the 41 in the network traffic. Bayesian Network and C

4.5 (a decision tree classifier) were used to detect DDoS attack in the network. Results

obtained shows that the detection accuracy remains the same while the overall efficiency

improved. Bolon-Canedo et al. [11] combined discretizers, filters and classifiers to improve

the classification performance by significantly reducing the feature set. This is applied to both

binary and multi-class classification problems using KDD Cup 99 benchmark dataset. A

supervised inductive learning approach called group method for data handling (GMDH) has

been proposed in [12] using two variant techniques; monolithic and ensemble-based. Filter

feature selection methods using IG, Gain Ratio and GMDH were used to rank features during

the pre-processing phase. Lin et al. [13] propose an anomaly intrusion detection that detects

new attacks using support vector machine (SVM), decision tree (DT) and simulated

annealing (SA). The best features were selected from the KDD’99 dataset using SVM and SA

to improve the classification accuracy of DT and SVM, to detect new attacks. Li et al. [16]

propose a gradual feature removal method that process dataset prior to combining cluster

method, ant colony algorithm and SVM to classify network traffic as either normal or

anomaly. Sindhu et al. [14] propose a wrapper method for feature selection to remove

irrelevant instances from a feature set to achieve higher detection accuracy using neuro tree.

A feature selection approach has also been proposed in [17] using Bayesian Network. NSL-

KDD dataset was used to evaluate the selected features and the result shows that the selected

feature decreases attack detection time and improved the classification accuracy as well as

the true positive rates. A recent work by Bhattacharya et al. [15] propose a multi-measure

multi-weight ranking approach that identifies important network features by combining

wrapper, filter and clustering methods to assign multiple weights to each feature.

Rough set feature selection approach has proven to be an efficient mathematical tool based on

upper and lower approximation. It presents equal classification capability with the minimal

subset. Olusola et al. [18] propose a rough set based feature selection method that selects

important features from input data using KDD ’99 dataset. Sengupta et al. [19] designed an

online intrusion detection system (IDS) using both rough set theory and Q-learning algorithm

to achieve a maximum classification algorithm to classify a data as either normal or anomaly

using NSL-KDD network traffic data. A fast attribute reduction algorithm based on rough set

剩余19页未读，继续阅读

xndh

粉丝: 0
资源: 11

云安全：集成多滤波器特征选择方法对抗DDoS攻击

Efficient-Facial-Feature-Learning-with-Wide-Ensemble-based-Convolutional-Neural-Networks:基于宽集合的卷积神经网络的高效面部特征学习

Ensemble-Pytorch:PyTorch的统一集成框架，可提高深度学习模型的性能和鲁棒性

Ensemble-based methods是什么

smote和Ensemble-based methods如何结合

smote和Ensemble-based methods结合有什么用，代码实现一下一下

smote和Ensemble-based methods如何结合，代码实现一下一下

Efficient facial feature learning with wide ensemble-based convolutional neural networks为什么可以用做情绪分类器

ensemble-xgboost

load(model_path = "./reg/paddlets-ensemble-model0",loaded_model0)有问题吗

paddle中load(model_path = "./reg/paddlets-ensemble-model0",program = loaded_model0)有问题吗

最新资源