ID 原文 译文
22805 在 LIVE 移动视频数据库上的实验结果表明，该文所提算法的结果与主观评价具有较好的一致性，能够准确反映人类对视频失真程度的视觉感知效果，可为实时在线调节信源码率和无线信道参数提供参考依据。 Experimental results in the LIVE mobile video database show that NMVQA is well consistent with subjective assessment results, and can reflect human subjective feeling well. NMVQA can evaluate the performance of real-time online adjustment of the source rate and wireless channel parameters.
22806 针对传统方法不能有效抽取维吾尔语事件因果关系的问题，该文提出一种基于双向 LSTM(Bidirectional Long Short-Term Memory, BiLSTM)的维吾尔语事件因果关系抽取方法。 Since the traditional events causal relation has the disadvantages of small recognition coverage, a method for causal relation extraction of Uyghur events is presented based on Bidirectional Long Short-Term Memory (BiLSTM) model.
22807 通过对维吾尔语语言以及事件因果关系特点的研究，提取出 10 项基于事件内部结构信息的特征； In order to make full use of the event structure information, 10 characteristics of the Uyghur events structure information are extracted based on the study of the events causal relationship and Uyghur language features;
22808 同时为充分利用事件语义信息，引入词嵌入作为 Bi LSTM的输入，提取事件句隐含的深层语义特征并利用批样规范化(Batch Normalization, BN)算法加速 Bi LSTM 的收敛； At the same time, the word embedding is introduced as the input of BiLSTM to extract the deep semantic features of the Uyghur events and Batch Normalization (BN) algorithm is usded to accelerate the convergence of BiLSTM.
22809 最后融合这两类特征作为 softmax 分类器的输入进而完成维吾尔语事件因果关系抽取。 Finally, concatenating these two kinds of features as the input of the softmax classifier to extract the Uyghur events causal relations. This method is used in the causal relation extraction of Uyghur events,
22810 实验结果表明，该方法用于维吾尔语事件因果关系的抽取准确率为  89.19%,  召回率为  83.19%, F 值为 86.09%，证明了该文提出的方法在维吾尔语事件因果关系抽取上的有效性。 The results show that the precision rate, the recall rate and F value can reach 89.19 %, 83.19% and 86.09 %, indicating the effectiveness and practicability of the method of causal relation extraction of Uyghur events.
22811 将随机矩阵的非渐近谱理论应用到协作频谱感知中，对接收信号样本协方差矩阵的最大特征值和最小特征值进行分析，该文提出一种精确的最大最小特征值差(Exact Maximum Minimum Eigenvalue Difference, EMMED)的协作感知算法。 The non-asymptotic spectral theory of random matrix is applied to cooperative spectrum sensing, the maximum eigenvalue and the minimum eigenvalue of the sampled signal covariance matrix are analyzed and an Exact Maximum Minimum Eigenvalues Difference (EMMED) algorithm is proposed.
22812 对于任意给定的协作用户个数 K 和采样点数 N，首先推导了最大最小特征值之差的精确概率密度函数(Probability Density Function, PDF)和累积分布函数(Cumulative Distribution Function, CDF)，然后利用该分布函数设计了所提算法的判决阈值。 For any given numbers of cooperative users K and sampling points N, the exact Probability Density Function (PDF) and Cumulative Distribution Function (CDF) of the difference between the maximum and minimum eigenvalues are derived. Then, an accurate decision threshold is designed by using the distribution function.
22813 理论分析表明，EMMED 算法的判决阈值较已有的渐进最大最小特征值差(Asymptotic Maximum Minimum Eigenvalue Difference, AMMED)检测更为精确，算法无需主用户信号特征并且能够对抗噪声不确定度影响。 Theoretical analysis shows, the EMMED algorithm has the characteristics that the decision threshold is more accurate than the existing Asymptotic Maximum Minimum Eigenvalue Difference (AMMED) algorithm, without the characteristics of the main user signal and not affected by noise uncertainty.
22814 仿真结果表明，存在噪声不确定度的感知环境下，EMMED 算法较已有的精确最大特征值(Exact Maximum Eigenvalue, EME)和 EMMER 等频谱感知算法具有更好的检测性能。 In addition, the simulation results show that the EMMED algorithm has better detection performance than the existing Exact Maximum Eigenvalue (EME) and EMMER algorithms in the real sensing environment with noisy uncertainty.