ID	原文	译文
4613	该方法不需要迭代计算，通过程序实现并与现有算法进行效率比较，结果表明所给出的新算法在计算周期为 p 2 的 q 元序列的 k–错线性复杂度方面效率明显更高。	Because the new method does not need iterative calculation and when it is implemented by program and compared with existing algorithms, the results show that the proposed new algorithm is significantly more efficient in calculating k-error linear complexity of q-ary sequence of period p 2 .
4614	面向知识服务过程中内容资源的智能化、知识化、精细化和重组化的碎片性管理需求。	In the process of knowledge service, in order to meet the fragmentation management needs of intellectualiza-tion, knowledge ability, refinement and reorganization content resources.
4615	深层分析并挖掘语义隐层知识、技术、经验与信息，突破已有传统文本到结构化查询语言（SQL）的语义分析技术瓶颈，提出基于预训练机制的自修正复杂语义分析方法 PT-Sem2SQL。	Through deep analysis and mining of semantic hidden knowledge, technology, experience, and information, it broke through the existing bottleneck of traditional se-mantic parsing technology from Text-to-SQL. The PT-Sem2SQL based on the pre-training mechanism was proposed.
4616	设计结合 Kullback-Leibler 差异技术的 MT-DNN 预训练机制，以加强上下文语义理解深度；	TheMT-DNN pre-training model mechanism combining Kullback-Leibler technology was designed to enhance the depth ofcontext semantic understanding.
4617	设计专有增强模块，捕获句内上下文语义信息的位置；	A proprietary enhancement module was designed that captured the location of contextualsemantic information within the sentence.
4618	并通过自修正方法优化生成模型的执行过程，以解决解码过程中的错误输出。	Optimize the execution process of the generated model by the self-correctingmethod to solve the error output during decoding.
4619	实验结果表明，PT-Sem2SQL 能够有效提高复杂语义的解析性能，准确度优于相关工作。	The experimental results show that PT-Sem2SQL can effectively im-prove the parsing performance of complex semantics, and its accuracy is better than related work.
4620	针对位置数据拥有者对数据使用者具有不同的信任程度时，会对使用者进行不同等级的划分并向其发布不同扰动程度的位置数据这一场景中的隐私泄露问题，提出了基于信息论中的互信息的隐私度量方法与保护问题。	A privacy metric based on mutual information was proposed to measure the privacy leakage occurred when location data owner trust data users at different levels and need to publish the distorted location data to each user accord-ing to her trust level, based on which an location privacy protection mechanism (LPPM)was generated to protect user'slocation privacy.
4621	此外，基于互信息提出了度量攻击者获取不同等级的扰动数据而对真实位置数据进行更精确的推断分析所造成的隐私泄露的方法。	In addition, based on mutual information, a metric was proposed to measure the privacy leakage caused by attackers obtaining different levels of distorted location data and then performing inference attack on the original loca-tion data more accurately.
4622	借鉴用于求解率失真函数的 Blahut-Arimoto 算法提出了多等级位置隐私保护机制。	Another privacy metric was also proposed to quantify the information leakage occurred in thescenario based on mutual information. In particular, the proposed privacy mechanism was designed by modifying Bla-hut-Arimoto algorithm in rate-distortion theory.