ID 原文 译文
56938 最后,本文总结了当前应用现状并对未来发展方向进行了展望. Finally, this paper summarizes the current application status and looksto future directions of development.
56939 随着谷歌知识图谱、DBpedia、微软Concept Graph、YAGO等众多知识图谱的不断出现,根据RDF来构建的知识表达体系越来越为人们所熟知. With the continued growth of various knowledge graphs, such as Google Knowledge Map, DBpedia,Microsoft Concept Graph, and YAGO, the knowledge representation system, constructed based on RDF, hasbecome more well-known.
56940 利用RDF三元组表达形式成为人们对现实世界中知识的基本描述方式,由于其结构简单、逻辑清晰,所以易于理解和实现,但也因为如此,当其面对现实中无比繁杂的知识和很多常识时,往往也无法做到对知识的认识面面俱到,知识图谱的构建过程注定会使其中包含的知识不具有完整性,即知识库无法包含全部的已知知识. The RDF triple format has become the basic description of knowledge in the realworld. Due to its simple structure and clear logic, it is easy to understand and implement. Nevertheless, whenfaced with extremely complicated knowledge and common sense, complete knowledge can become difficult todescribe. The construction process of knowledge graphs is bound to lead to incomplete knowledge contained inthe graphs.
56941 此时知识库补全技术在应对此种情形时就显得尤为重要,任何现有的知识图谱都需要通过补全来不断完善知识本身,甚至可以推理出新的知识. At this point, the knowledge-based completion technology is particularly important for managingsuch situations. Any existing knowledge graph must be improved continuously through completion technology andnewly inferred knowledge.
56942 本文从知识图谱构建过程出发,将知识图谱补全问题分为概念补全和实例补全两个层次: Beginning with the construction of a knowledge graph, this paper divides the problemof knowledge graph completion into two levels: concept completion and instance completion.
56943 (1)概念补全层次主要针对实体类型补全问题,按照基于描述逻辑的逻辑推理机制、基于传统机器学习的类型推理机制和基于表示学习的类型推理机制等3个发展阶段展开描述; (1) The conceptcompletion level primarily focuses on the completion of entity types. It is described in terms of three developmentstages: a logical reasoning mechanism, based on description logic, a type inference mechanism, based on traditionalmachine learning, and a type inference mechanism, based on representation learning.
56944 (2)实例补全层次又可以分为RDF三元组补全和新实例发现两个方面,本文主要针对RDF三元组补全问题沿着统计关系学习、基于随机游走的概率学习和知识表示学习等发展阶段来阐述实体补全或关系补全的方法. (2) The instance completionlevel can be further divided into an RDF triple completion and new instance discovery. This paper focuses onRDF triples completion learning, which includes entity completion or relationship completion and is described inthree development stages, such as statistical relational learning, probability learning based on random walks, andknowledge representation learning.
56945 通过对以上大规模知识图谱补全技术研究历程、发展现状和最新进展的回顾与探讨,最后提出了未来该技术需要应对的挑战和相关方向的发展前景. Through the review and discussion of the research process, the developmentstatus, and the latest progress in the above-mentioned large-scale knowledge graph completion, we present thechallenges that the technology will face and the development prospects of future work.
56946 现有DNS权威服务器处理DNS请求及响应报文依赖软件网络协议栈, CPU资源占用率高、开销大,处理性能受限. The existing, authoritative DNS servers process DNS requests and response packets depending onthe software network protocol stack, which use DNS resources and have high overhead and limited processingperformance.
56947 本文基于SmartNIC架构对DNS权威服务器的功能进行卸载加速,提出并设计了高性能DNS权威查询响应流水线PHDR Pipe (perfect Hash DNS response pipeline),基于完美哈希(perfect Hash)实现对区文件的预先处理,避免哈希冲突导致的多次访存,降低流水线最坏情况下处理延迟,从而有效提升系统吞吐率并降低响应延迟. Based on the SmartNIC architecture, this paper quickens the unloading of DNS authority serverand proposes and designs a high-performance DNS authority query response pipeline PHDR Pipe (Perfect HashDNS Response Pipeline), which realizes the preprocessing of region files based on perfect Hash. To avoid multiplememory access, caused by Hash collision, and reduce the processing delay in the worst case of a pipeline, thesystem throughput is effectively improved and reduces the response delay.