期望与现实:评估实践中的入侵检测系统
Expectations Versus Reality: Evaluating Intrusion Detection Systems in Practice
AI总结:
本文通过实证比较多种入侵检测系统,揭示其表现受攻击类型、复杂度和网络环境影响,指出深度神经网络虽平均 F1 最高但并非始终最优,并讨论了实际使用中的困难。
摘要:
本文对近年来的入侵检测系统(IDS)进行了实证比较,旨在提供客观对比,帮助用户根据自身需求选择最合适的解决方案。我们的结果表明,没有单一方案是最优的,其表现取决于外部变量,例如攻击类型、复杂度以及数据集中的网络环境。例如,BoT_IoT 和 Stratosphere IoT 数据集都包含与物联网相关的攻击,但在使用 BoT_IoT 数据集测试时,深度神经网络表现最佳;而在使用 Stratosphere IoT 数据集测试时,HELAD 表现最佳。因此,尽管我们发现深度神经网络方案在测试数据集上取得了最高的平均 F1 分数,但它并非始终是表现最好的方案。我们进一步讨论了从文献和项目仓库中使用 IDS 所面临的困难,这些困难使得在 IDS 选择方面难以得出确定性结论。
英文摘要:
Our paper provides empirical comparisons between recent IDSs to provide an objective comparison between them to help users choose the most appropriate solution based on their requirements. Our results show that no one solution is the best, but is dependent on external variables such as the types of attacks, complexity, and network environment in the dataset. For example, BoT_IoT and Stratosphere IoT datasets both capture IoT-related attacks, but the deep neural network performed the best when tested using the BoT_IoT dataset while HELAD performed the best when tested using the Stratosphere IoT dataset. So although we found that a deep neural network solution had the highest average F1 scores on tested datasets, it is not always the best-performing one. We further discuss difficulties in using IDS from literature and project repositories, which complicated drawing definitive conclusions regarding IDS selection.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。