During the second week, Linh and I dug into RDF/Turtle format and SPARQL syntax for testing the existing incomplete Turtle file (refer to Prefix rdf, rdfs, provone, owl and self-defined yw) and relevant queries. I successfully installed Apahe-Jena on my Windows machine and tested the five existing simple SPARQL queries using arq command. When taking a close look at the Turtle file, I found some syntax and semantic errors, such as inconsistent names and incomplete ports descriptions. Then I manually corrected those errors based on my perception. With the updated version of Turtle file, I translated most existing xsb queries into SPARQL queries and ran the queries in Apache-Jena as well. Simultaneously, Linh did the translation and testing work using Virtuoso on her laptop. During the translation work, I felt that aggregation related queries are harder to be written in SPARQL than written in Datalog/Prolog or SQL.
Besides, I installed msys2 on Windows 10 to execute make command and to test the yw-prototypes tool again. Although some problems still could not be fixed, I learned about another method to utilize Linux-like terminal environment on Windows machine. Moreover, I further explored the ProvONE data model, especially for the Workflow Representation part. Then, I reviewed the Prolog/Datalog model and recon facts in details and summarized all the structures, comparing with the RDF semantic structure in the temporal Turtle file.
For the following weeks, I am going to read some papers about the expressive power and limitation of SPARQL language, comparing with my experience when doing the translation work. Additionally, I will set about improving the Turtle file based on the needs of SPARQL queries and ProvONE data model.
Hui