Hello everyone, this Xiaoliang Jiang from project 1. This is the third week for my internship. For this week, I used several days to keep learning how to create a R extension, and then based on that, I have created a “prototype” function to output a RDF file by using the annotations created by “strcode” package.
For the first two days in this week, I kept learning how to create a R package (R extension) by reading some official help document and relative papers. I also discussed my working plan with my mentor online to confirm my tasks for this week. According to the 9-week working schedule, in this week, my primary task is trying to find out at least one way to summarize the semantics in annotated scripts and output them in RDF file.
For the rest time of this week, I created several functions to generate a RDF file from annotated R file. At the beginning, I spend a lot of time to understand the logic in depth about the “sum_str” function in “strcode”. I tried to create a new function called sum_rdf to generate RDF file. But due to my limited understanding on R package, there are always some problems in my edited package. So, at last, I decided to add a new argument into the original function as a new option for it to create a RDF file. For the original function, if I run sum_str() function, it will create a summary of annotated information. After I added an additional input called “rdf”, if “rdf=TRUE”, the function will create an RDF file instead of a summary of annotations by annotated information. Although it is only a prototype which does not contains any header, only has limited ProvONE words, and has many other problems, it at least could be a successful attempt and a positive way to generate RDF file. In addition, it provided several possible improvement options, like adding more different types of schema by extending word list.
For the next week, I think I could try to revise this prototype function, and add more information such as prefix headers into the RDF file.