相似重复记录检测研究与发展动态的知识图谱分析  

KNOWLEDGE MAP ANALYSIS OF RESEARCH AND DEVELOPMENT TRENDS IN SIMILAR DUPLICATE RECORD DETECTION

作　　者：顾晴 董永权[1] 胡杨 Gu Qing;Dong Yongquan;Hu Yang(School of Wisdom Education,Jiangsu Normal University,Xuzhou 221116,Jiangsu,China)

机构地区：[1]江苏师范大学智慧教育学院,江苏徐州221116

出　　处：《计算机应用与软件》2022年第3期1-7,95,共8页Computer Applications and Software

基　　金：国家自然科学基金项目(61872168);江苏省研究生科研与创新计划省级项目(KYCX20_2396)。

摘　　要：在大数据环境下,数据库中的记录数量呈指数上升,如何高效率地检测出相似重复记录是数据清洗的关键点和提高数据质量的首要任务。近十年国内外相似重复记录检测方法又涌现出相当多的高水平成果,迫切需要对新的文献加以归纳梳理。以2008—2019年的国内外相似重复记录检测相关文献为研究样本,结合社会网络和知识图谱对其发文量、核心机构、作者合作群、研究热点和研究趋势进行分析。分析发现,作者合作结构整体上较松散,相似重复记录各类检测方式的集成、应用领域的扩展和通用框架的研究成为热点,缺失数据值的处理、多数据源的识别、大数据量的分块处理成为相似重复记录领域的挑战。In the big data environment,the number of records in the database is increasing exponentially.How to efficiently detect similar duplicate records is the key point of data cleaning and the primary task of improving data quality.In the past ten years,there have been quite a few high-level achievements in the similar duplicate record detection methods at home and abroad.It is urgent to summarize the new literature.We took the relevant literature of similar duplicate records at home and abroad in 2008-2019 as research samples.And combing with social networks and knowledge maps,we analyzed the number of papers,core institutions,author cooperative groups,research hotspots and research trends.The study finds that the author’s cooperative structure is loose overall,and the integration of various types of detection methods,the extension of application fields and the research of general frameworks have become hot spots.The processing of missing data values,the identification of multiple data sources,and the partitioning of large data volumes become challenges in the field of similar duplicate records.

关 键 词：相似重复记录检测 知识合作 研究热点 研究趋势 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]