Abstract: Nowadays linked data popularity increases along with information description in the form of RDFtriplets. Efficient implementation of users’ interaction with these kinds of data requires studying communication procedures with triple stores. One of the main difficulties, that are currently unsolved, is the complexity of dynamic querying procedures. We try to deal with this issue by creating query patterns and decreasing query complexity. A unified search interface is developed, which enables visual querying the triple stores implemented through OpenLink? Virtuoso universal server. Visual queries are automatically converted into SPARQL query language, which is used for accessing the triple stores. After the query is executed, a user gets the desired context with triplets according to constraints for predicates and objects. Also the formal model is developed, which mathematically describes linked data based on partially defined object schemas. Existing search model for distributed environments is improved. The developed practical implementation for medical institution is under stage of manufacturing application.

Keywords: linked data, RDF data management, query pattern, triple store, semantic application.

ACM Classification Keywords: E.2 DATA STORAGE REPRESENTATIONS (Linked representations), I.2.4 Knowledge Representation Formalisms and Methods (Representation languages)

Link:

PECULIARITIES OF LINKED DATA PROCESSING IN SEMANTIC APPLICATIONS

Sergey Shcherbak, Ilona Galushka, Sergey Soloshich, Valeriy Zavgorodniy

http://www.foibg.com/ijima/vol02/ijima02-02-p07.pdf