基于Hadoop的大规模RDF语义数据应用平台  

Application Platform for Large-scale Semantic RDF Data based on Hadoop

作　　者：肖宝[1] 李璞[2] 胡文君[1] 韦丽娜[3] XIAO Bao LI Pu HU Wenjun Wei Lina(School of Electronics and Information Engineering, Qinzhou University, Qinzhou 535011, China Software Engineering College, Zhengzhou University of Light Industry, Zhengzhou 450000, China School of Foreign Language, Guangxi University for Nationalities, Nanning 530006, China)

机构地区：[1]钦州学院电子与信息工程学院,广西钦州535011 [2]郑州轻工业学院软件学院,河南郑州450000 [3]广西民族大学外国语学院,广西南宁510631

出　　处：《钦州学院学报》2017年第1期12-17,共6页Journal of Qinzhou University

基　　金：广西高校中青年教师基础能力提升项目:基于Wikipedia的大规模Web文本分类的研究(KY2016-LX431)

摘　　要：随着语义网技术不断发展,传统推理引擎技术对大规模RDF(Resource Description Framework)数据的高效存储和语义查询存在着计算性能差和扩展能力不足等问题。针对这些问题,引入大数据处理技术Hadoop以及No SQL存储技术,构造一个大规模RDF语义数据应用平台,并且设计了基于SPARQL(Simple Protocol and RDF Query Language)查询技术的节点资源扩展算法。平台充分利用MapReduce技术提高了数据的存储和查询性能。实验以DBpedia的语义数据为案例,验证了平台的可行性和有效性。With the development of semantic web, conventional rule-based engines meet the bottleneck in computing performance and scalability for storage and semantic query. To address these problems, application platform based on Hadoop and NoSql for large-scale semantic RDF （Resource Description Framework） data was presented. Algorithm based on SPARQL （Simple Pro- tocol and RDF Query Language） was also designed for extending resource node. Performance of storage and query was improved by taking advantage of MapReduce. The feasibility and effectiveness of the presented platform were proven through the case of DBpedia semantic data set.

关 键 词：RDF HADOOP 大规模语义数据 MAPREDUCE 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]