面向跨语言家谱服务的多源关联数据匹配研究——上海图书馆开放数据应用比赛作品Learn Chinese Surnames

Matching Linked Data for Cross-Lingual Genealogical Services——Learn Chinese Surnames in Shanghai Library Open Data Application Contest

随着越来越多非英文关联数据集的发布,语言差异成了数据万维网中资源间相互链接的障碍。对于包括中文家谱在内的文化遗产资源,以关联数据技术为基础的跨语言信息服务有助于保障资源的获取,促进国际交流。此文介绍了关联数据中跨语言本体匹配的概念、代表项目和在文化遗产领域的相关实践,并以2016年上海图书馆开放数据应用开发竞赛作品Learn Chinese Surnames作为案例,展示了用关联数据的消费技术实现中文家谱数据与DBpedia、GeoNames、Wiktionary之间的跨语言匹配,以丰富中文家谱中的相关英文描述的实践,并总结了开发关联数据应用的经验。该研究有助于消除关联数据集之间的语言障碍,实现跨语言的家谱信息服务。

With the growth of non-English linked datasets, discrepancy in language has become a major obstacle for cross-lingual linking of resources in the Web of Data. For cultural heritage resources including Chinese genealogies, cross-lingual information services based on linked data technologies can support access of these resources and thus promote international scholarly communication. In this article, we present the idea of Cross-Lingual Ontology Matching and representative projects especially in the Cultural Heritage domain. The work, Learn Chinese Surnames, winning the Shanghai Library Open Data Application Contest in 2016, is then introduced as a case study to demonstrate the practice of consuming Linked Data to estab-lish Cross-Lingual Matching between Chinese Genealogical Data to DBpedia, GeoNames and Wiktionary. The experiences of developing linked data applications has also been summarized. The research is useful to overcome the language barriers among linked datasets and to realize cross-lingual genealogical information services.