ID	原文	译文
3583	所提安全模型基于区块链技术解决传统贸易系统中的中心化集权的问题，采用基于同态认证的公开审计技术实现了无管理者的安全背书、背书密钥更新及贸易的高效验证，引入基于信誉的激励机制保证了贸易的及时性。	The proposed security model realized decentralization based on the blockchain technology, by using homomorphic au-thentication-based public auditing to achieve the secure endorsement and key updating for non-manager groups and effi-cient verification for transaction information respectively, introducing the reputation-based incentive mechanism to en-sure the timeliness of the transaction.
3584	最后，通过安全证明与性能分析表明了所提安全模型的安全性和可靠性，通信和计算开销均低于 IPANM。	Finally, the security and reliability of the proposed security model were demon-strated by security proof and performance analysis, and the communication and computing costs are both lower thanIPANM.
3585	针对基于图的多标签特征选择方法忽略图拉普拉斯矩阵的动态变化，且利用逻辑标签来指导特征选择过程而丢失标签信息等问题，提出了一种基于动态图拉普拉斯矩阵和实值标签的多标签特征选择方法。	In view of the problems that graph-based multi-label feature selection methods ignore the dynamic change ofgraph Laplacian matrix, as well as such methods employ logical-value labels to guide feature selection process and loseslabel information, a multi-label feature selection method based on both dynamic graph Laplacian matrix and real-valuelabels was proposed.
3586	该方法利用特征矩阵的稳健低维空间构造动态图拉普拉斯矩阵，并利用该稳健低维空间作为实值标签空间，进一步使用流形约束和非负约束将逻辑标签转化为实值标签，以此来解决上述问题。	The robust low-dimensional space of feature matrix was used to construct a dynamic graph Lapla-cian matrix, and the robust low-dimensional space was used as the real-value label space. Furthermore, manifold andnon-negative constraints were adopted to transform logical labels into real-valued labels to address the issues mentioned above.
3587	所提方法与 3 种多标签特征选择方法在 9 个多标签基准数据集上进行了对比实验，实验结果表明，所提多标签特征选择方法可得到高质量的特征子集，并且能获得很好的分类表现。	The proposed method was compared to three multi-label feature selection methods on nine multi-label benchmarkdata sets in experiments. The experimental results demonstrate that the proposed multi-label feature selection method canobtain the higher quality feature subset and achieve good classification performance.
3588	针对极化码盲识别问题，首先证明了能表征实际极化码码长、码率关系的定理 1 和定理 2 及区别冻结比特位和信息比特位的定理 3。	In order to solve the problem of the blind recognition of polar codes, the theorem 1 and theorem 2 were provedfirstly, which reflects the relationship between length and rate of actual polar codes, and then theorem 3 which could dis-tinguish frozen bit and information bit positions was also proved.
3589	基于这 3 个定理，通过遍历可能的码长值，构建了码字矩阵和克罗内克矩阵，然后遍历信息比特位，检测码字空间与疑似对偶空间的校验关系。	Based on these three theorems, the codewords matrixesand Kronecker matrixes were constructed by traversing the possible code length values. Then the information bits weretraversed to detect the check relationship between the codewords and the suspected dual space.
3590	为了检测校验关系，引入了对数似然比概念，基于其统计特性和最优准则，估计出该遍历码长下的码率及信息比特位置，最终完成参数的识别。	In order to detect thecheck relationship, log likelihood ratio was introduced, based on its characteristics and optimal criteria, the code rate andinformation bit positions were estimated.
3591	仿真结果表明，3个定理的结论与仿真结果一致，且算法具有较强的容错性，在信噪比为 6.5 dB、码长为 1 024 条件下，参数识别率能够达到 98%以上。	The simulation results show that the conclusions of the three theorems are con-sistent with the results. At the same time, the proposed algorithm has a strong error tolerance. Under 6.5 dB and codelength of 1024, the rate of recognition can reach more than 98%.
3592	针对现有基于生成文本和社交关系的联合位置推断方法对社交媒体中异质数据间的位置关联性挖掘不充分的问题，提出了一种基于多种提及关系的社交媒体用户位置推断方法。	Aiming at the problem that the existing joint user geolocalization methods based on social media text and so-cial relationships do not sufficiently mine the location correlation between heterogeneous data in social media, a socialmedia user geolocalization method based on multiple mention relationships was proposed.