Linked Data Technology And Its Application In Biomedicine

Semantic Web makes it possible to share and reuse data and knowledge freely. Linked data is essential to actually connect the Semantic Web, and it is widely used in government data, multimedia and biomedicine. The need of data integration in biomedicine is increasing due to its requirement of cross-discipline cooperation. Tra-ditional data integration techniques such as data warehouse are centralized local inte-gration and lack of scalability. The distributed nature of Semantic Web is capable of achieving data integration globally. Ontology matching and instance linking are nec-essary to integrate linked data.Plenty of datasets are available on Linked Open Data cloud. However, there are still a great amount of legacy data remaining in relational database. The key technique to integrate legacy data is to find mappings between relational database schemas and ontologies.Expressing query intention in SPARQL and memorizing domain ontology are re-quired to retrieve integrated linked data, which hinder users from accessing linked data. While keyword query is ambiguous. Controlled natural language is a tradeoff between formal query language and keyword query. The main contributions of this paper are listed as following:1. A system named BioSearch is designed and built, which integrated several hetero-geneous datasets.2. An approach is proposed to create mappings between relational database schemas and ontologies. The simple mappings are harnessed to integrate legacy data in rela-tional database. And the complex mappings are used in semantic recommendations.3. A controlled natural language named SemLabel is designed to facilitate accessing linked data.