ID 原文 译文
44316 基于原始数据与经验特征工程相结合的思想提出了一种混合结构深度神经网络，结合了卷积神经网络与多层感知机，分别处理文本与统计信息。 Then, a hybrid neural network was proposed based on the combination of raw data and empirical feature engineering. It combined convolutional neural network (CNN) and multi-layer perceptron (MLP) to process text and statistical information.
44317 与传统机器学习算法(如 SVM)相比，所提方法效果提升明显，F 1 值可达 99.38%，且具有更低的时间代价。 The effect of the model was significantly improved compared with traditional machine learning algorithms (e.g. SVM). The F 1 value reached 99.38% and had a lower time complexity.
44318 标注了一套由 45 万余条恶意流量和 2 000 万余条非恶意流量组成的数据集。 At the same time, a data set consisting of more than 450 000 malicious traffic and more than 20 million non-malicious traffic was created.
44319 并依据模型设计了一套原型系统，精确率达到了 98.1%~99.99%，召回率达到了 97.2%~99.5%，应用在真实网络环境中效果优异。 In addition,prototype system based on model was designed with detection precision of 98.1%~99.99% and recall rate of 97.2%~99.5%. The application is excellent in real network environment.
44320 毫米波的定向天线和波束赋形技术使得多条数据流之间的并行传输成为可能。 The directional antennas and beam-forming techniques in millimeter wave (mm Wave) bands are used to make concurrent transmission between multiple flows become possible.
44321 但并行传输也可能造成更高的多用户干扰。 However, higher mutual interference may be caused by concurrent transmission.
44322 因此在时隙资源有限且请求传输的数据流数较多时，针对如何对数据流进行有效的并行调度以提高网络性能，提出了一种基于网络全局冲突图的并行调度算法（CB-STDMA）。 Therefore, when the time slots were limited and the number of data flows was large, how to schedule the concurrent flows efficiently was solved by proposed algorithm.
44323 为保证用户的服务质量（QoS）需求，该算法以最大化网络中满足 QoS 需求的流数为目标，考虑了流之间的冲突干扰，并且优先对所用时隙数较少的流进行调度来节省资源。 The contention graph based spatial-time division multiple access (CB-STDMA) concurrent scheduling algorithm guaranteed the quality of service (QoS) of users, and aimed at maximizing the number of flows with their QoS requirements satisfied. It considered the interference between different flows, and a higher priority was given to the flow with fewer time slots requirement.
44324 通过仿真验证了该算法性能的优势。 Extensive simulations demonstrated that the proposed CB-STDMA algorithm had some advantages compared with the existing algorithms.
44325 相较于传统综合字典学习方法，非参数贝叶斯方法具有显著优势，但其对图像结构全局相似性和变异性的表示能力仍有较大提升空间。 Though non-parametric Bayesian methods possesses significant superiority with respect to traditional comprehensive dictionary learning methods, there is room for improvement of this method as it needs more consideration over the structural similarity and variability of images.