ID 原文 译文
44556 结果表明，混合分类算法模型（CCAM）对 DNS 隧道的检测能达到 90%精度值。 Experimental results showed that CCAM has the high performance compared to the other classifiers in terms of detecting DNS tunneling by achieving 90% of F-measure.
44557 随着 IPv4 地址的耗尽，IPv6 将被广泛部署使用。 With the gradual exhaustion of IPv4, IPv6 will be widely used.
44558 在互联网流量分类领域，目前大多研究对象为 IPv4 流量。 In Internet traffic classification field, most re-searches mainly focus on the IPv4 traffic.
44559 采集校园网 IPv6 流量数据，开展数据预处理工作，为解决流量数据面临的概念漂移问题，改进流式机器学习算法（VFDT）。 Campus IPv6 network traffic data was captured, carried out pre-process on these raw traffic data.
44560 为每个类别搜索与之相关的特征集合，利用各特征集合建立子空间， It improves the streaming learning algorithm VFDT to address the concept drift problem faced by net-work traffic data. The method searches a correlated feature subset for each class, and each feature subset builds a subspace.
44561 在每个子空间利用流式机器学习算法建立分类模型，在分类过程中，不断学习新标记流量数据，改善分类模型。 And then the streaming learning algorithm is used train a model, and the model is continuously updated on the samples in each sub space.
44562 未知流量数据的类别预测为多个分类模型的投票结果。 The weighted voting is used for predicting the label of unknown flow samples.
44563 实验结果表明，改进的算法能改善流式学习方法在 IPv6 网络流量数据上的分类性能。 The experimental results show that the method can improve the performance of the stream learning method on IPv6 traffic data.
44564 超高密度无线局域网，特指在用户密度超高的场景中为用户提供实时服务的大规模无线局域网。 Ultra-high density wireless network is a large scale wireless network that provides real-time service in an extremely high user density circumstance.
44565 在超高密度环境中，由同频干扰引发的冲突远高于传统场景，因此传统的信噪比及信道利用率无法准确反映网络的传输质量。 In ultra-high density environment, SINR takes the place of SNR to reflect the transport quantity of the WLAN because of more co-channel interference.