[目的/意义]对图书馆数据形态演变情况进行探讨，以了解各种形态下图书馆数据研究的重点及未来研究方向。[方法/过程]在考察OCLC的4个数据科学项目及其反映出的3种数据表现形式的基础上，将图书馆数据形态分为原生形态、聚合形态、扩展形态、关联形态、知识形态等5种类型。[结果/结论]不同的图书馆数据形态有着不同的数据加工深度和广度。可通过多源聚合、内生与外联、关联数据化、知识三元组抽取与融合等途径实现5种数据形态之间的有序演变。
[Purpose/significance] Discussion on the evolution of library data forms is helpful to understand research emphases on library data in various forms, as well as the future research direction.[Method/process] Based on investigations on four OCLC data science projects and three forms of data representation reflected in them, forms of library data are divided into five types, including original data form, aggregation form, extended data form, linked data form and knowledge form.[Result/conclusion] The processing depth and breadth of library data is gradually strengthened with its form evolution. The orderly evolution between these forms can be achieved through multi-source polymerization, internal generation and external association, converting library data into linked data, extraction and fusion of knowledge triples.
[1] 白广思. 数字图书馆大数据分类研究[J]. 图书馆学研究,2016(3):53-57.
[2] 郭秋萍,王全兰. 一种基于Web挖掘的图书馆服务推荐模型及其算法研究[J]. 图书馆杂志,2010(6):53-56.
[3] BAKER T,BERMES E,COYLE K,et al.Library linked data incubator group final report[EB/OL].[2016-07-05]. http://www.w3.org/2005/Incubator/lld/XGR-lld-20111025/.
[4] ISAAC A,WAITES W,YOUNG J,et al.Library linked data incubator group:datasets, value vocabularies, and metadata element sets[EB/OL].[2016-07-05]. http://www.w3.org/2005/Incubator/lld/XGR-lld-vocabdataset-20111025/.
[5] 曾蕾. 关联的图书馆数据[EB/OL].[2016-07-05]. http://wenku.baidu.com/view/9b26ca136edb6f1aff001f13.html.
[6] 包冬梅,范颖捷,李鸣. 高校图书馆数据治理及其框架[J]. 图书情报工作,2015,59(18):134-141.
[7] CHEN H, DOTY P, MOLLMAN C, et al. Library assessment and data analytics in the big data era:practice and policies[EB/OL].[2016-12-25]. http://onlinelibrary.wiley.com/doi/10.1002/pra2.2015.14505201002/full.
[8] COYLE K. Library data in the web world[EB/OL].[2016-12-23]. https://journals.ala.org/index.php/ltr/article/view/4673/5546.
[9] COYLE K. Chapter 2:changing the nature of library data[EB/OL].[2016-12-23]. https://journals.ala.org/index.php/ltr/article/view/4629/5475.
[10] FOURIE L. Surviving digital darwinism in the library[EB/OL].[2016-12-25]. http://uir.unisa.ac.za/handle/10500/21165.
[11] WESTRUM A, REKKAVIK A, TALLERÅS K. Improving the presentation of library data using FRBR and linked data[EB/OL].[2016-12-23]. http://journal.code4lib.org/articles/6424.
[12] FONS T, PENKA J, WALLIS R. OCLC's linked data initiative:using Schema.org to make library data relevant on the web[EB/OL].[2016-12-23]. http://www.niso.org/apps/group_public/download.php/9408/IP_Fons-etal_OCLC_isqv24no2-3.pdf.
[13] FONS T. Making library collections discoverable on the web[EB/OL].[2016-12-23]. https://www.oclc.org/content/dam/oclc/events/2015/CI_SFPL_PPTS/EntitiesAndWeb_SFPL2015.pdf.
[14] OCLC. VIAF contributor institutions[EB/OL].[2016-07-15]. http://www.oclc.org/viaf/contributors.en.html.
[15] OCLC. Inside worldcat[EB/OL].[2016-07-10]. http://www.oclc.org/worldcat/inside-worldcat.en.html.
[16] DONG X L, GABRILOVICH E, HEITZ G. Knowledge vault:a web-scale approach to probabilistic knowledge fusion[EB/OL].[2016-07-12]. https://www.cs.ubc.ca/~murphyk/Papers/kv-kdd14.pdf.
[17] 在线汉语词典[EB/OL].[2016-07-15]. http://cd.diyifanwen.com/hanyucidian/X/09330015357438728.htm.
[18] 刘炜. 关联数据:概念、技术及应用展望[J]. 大学图书馆学报,2011(2):5-12.
[19] MALMSTEN M. Exposing library data as linked data[EB/OL].[2016-12-23]. http://wtlab.um.ac.ir/images/e-library/linked_data/other/Exposing%20Library%20Data%20as%20Linked%20Data.pdf.
[20] Index (data.bnf.fr)[EB/OL].[2016-12-23]. http://data.bnf.fr/en/.
[21] RINDFLESCH T, ARONSON A. Semantic processing for enhanced access to biomedical knowledge[EB/OL].[2016-06-10]. http://skr.nlm.nih.gov/papers/references/semwebapp.5a.pdf.
