Skip to content

stream-project/WP1R

Repository files navigation

WP1R

STREAM WP1: MatVoc Ontology Entwicklung

Performance Comparison of uploading and querying RDF Triples of following repositories:

       1. GraphDB
       2. RDFox
       3. Allegrograph
       4. Virtuoso

In the Experiment folder we have three sub-folders (datasets, scripts and results) and two files(Queries.txt and comparison of uploading and querying triples), file "Queries.txt" has all the queries that are used for comparing time of uploading and querying rdf triples of the above mentioned repositories and the file named "comparison of uploading and querying triples" has uploading and querying time comparison of the four repositories used in our work.

Sub-folder "datasets" consists of all the datasets used in our work, it has Four RDF files namely:

      1. person.ttl: which contains details of the person's like first and last names.
      2. company.ttl: which contains details of the company.
      3. relation.ttl: which defines the relation between person and the company.
      4. company_relation.ttl: it is the combined version of relation and company datasets.

Sub-folder "scripts" has two folders in it:

      1. upload_script: which contains four python scripts that we have used to upload the triples into the repositories and to measure the uploading time.
      2. query_script: It also contains python scripts that we have used to query the above mentioned repositories. 
      
Note: Inside the query_script, there is another folder named "rdfox" which contains python script that we have 
used to query rdfox and also it contains one text file which has three queries that we have used with 
Jena Fuseki Server. 

Sub-folder "results" contains all the results of the queries that we have performed.

Experiment_v2

Experiment_v2 folder contains results of querying and uploading triples of two graph databases: GraphDB and RDFox, here we have combined all the datasets that we have seen above into a single dataset and uploaded into GraphDB and RDFox and then query it with the same queries that we have used above (NOTE: query1 and query8 have been modified; query1 has a different format and in query 8 we are querying all the results, so basically we have removed the LIMIT which was 500 before in query 8)

In the experiment_v2 folder there are two sub-folders and two files. file "queries.txt" has all the nine queries that we have used and file "combined_dataset" has triples uploading and querying result table.

The two sub-folders; "results" and "scripts" contains the results of the queries and the python scripts that we have used for uploading and querying triples respectively.