ID	原文	译文
3163	将威胁类型、严重程度、置信度、攻击频度、传播方式等策略匹配条件进行排序，提出了带模糊算子的区间决策图构造算法，设计了面向威胁处置的策略动态匹配算法，实验证明了其有效性。	Considering threat type, threat level, attack frequency and propagation mode, an algorithm with fuzzy operator was designed to construct interval decision diagram.Further, a fuzzy match algorithm was proposed to quickly select response policies. Experimental results show the effi-ciency of the proposed approach.
3164	传统的网络异常流量检测方法往往存在特征选择差与泛化能力较弱等缺陷，导致检测精度较低。	Problems such as a difficulty in feature selection and poor generalization ability were prone to occur when tra-ditional method was exploited to detect abnormal network traffic.
3165	为此，提出了一种基于长短记忆网络（LSTM）与改进残差神经网络优化的异常流量检测方法。	Therefore, an abnormal traffic detection method basedon the long short term memory network (LSTM) and improved residual neural network optimization was proposed.
3166	首先分析网络流量特征，通过预处理来降低网络流量特征值的差异性；	Firstly, the features and attributes of network traffic were analyzed, and the variability of the feature values was reducedby preprocessing of network traffic.
3167	然后设计了一种三层堆叠 LSTM 网络来提取不同深度的网络流量特征；	Then, a three-layer stacked LSTM network was designed to extract network trafficfeatures of different depths. Moreover, the problem of weak adaptability of feature extraction was solved.
3168	最后设计了一种带跳跃连接线的改进残差神经网络对 LSTM 进行优化，改善了深度神经网络中的过拟合与梯度消失等缺点，从而提高网络异常流量检测的准确率。	Finally, an im-proved residual neural network with skipping connecting line was designed to optimize the LSTM. The defects of deepneural network such as overfitting and gradient vanishing were optimized. The accuracy of abnormal traffic detection wasimproved.
3169	实验表明，所提方法具有较高的训练准确率，数据处理的可视性效果较好，二分类和多分类下的分类准确率分别为 92.3%和 89.3%。	Experimental results show that the proposed method has higher training accuracy and better visibility of dataprocessing. The classification accuracy rates under two classifications and multiple classifications are 92.3% and 89.3%.
3170	与当前入侵检测方法相比，所提方法在精确率、召回率等参数最优时具有最低的误报率。	It has the lowest false positive rate when the parameters such as precision rate and recall rate are optimal.
3171	在数据样本在遭到破坏时具有较强的稳健性，同时也具有较好的泛化能力。	Moreover, i thas strong robustness when the sample is destroyed. Furthermore, better generalization ability can be achieved.
3172	针对智能检测模型的性能受限于原始数据（特征）表达能力的问题，设计了一种残差网络结构 ResNet-32用于挖掘区块链交易特征间隐含的关联关系，自动学习包含丰富语义信息的高层抽象特征。	Aiming at the problem that the performance of intelligent detection models was limited by the representationability of original data (features), a residual network structure ResNet-32 was designed to automatically mine the intricateassociation relationship between original features, so as to actively learn the high-level abstract features with rich seman-tic information.