converting excel to rdf - ontology- querying using SPARQL

[varun ramireddy]

Hi all, 

I am trying to do the below using TopBraid Composer Maestro edition :

1) convert an excel file to RDF triples (output format required as .ttl)

2) create an Ontology so that relationships between columns can be mapped

3) followed by the creation of an ontology and triples (by using excel data), I would like to query the triples using SPARQL. 

Being new to the TopBraid Composer Maestro software, any help/ advice on the above is very much appreciated. 

Kind regards, 

Varun

[Irene Polikoff]

Logically speaking, you need an ontology before you convert data into triples. Otherwise, the best a software can do is assume that there is a single class and each column is a property with literal values. In other words, no relationships.

In any case, have you tried to just open the file in TBC? It should ask you if you already have an ontology. If not, it will create properties using the above assumption. The ontology must have annotations to help transform data in a more “custom” way.

Go to Help->TopBraid Composer > Importing Data Sources > Import external information > Import Spreadsheets >Handling Table documents with Semantic Tables to learn more.

Once your data is converted, query it with SPARQL.

Another option:

If you have TBC Maestro, you also have a local version of TopBraid EDG - go there from the TopBraid Applications menu in TBC.

1. Create an ontology in EDG

2. Create classes reflected in your data. If you are talking about relationships, you are likely to have more than one class. If your relationships are between members of the same class, then you only need one class.

3. Populate class properties from the spreadsheet using Derive property shapes from spreadsheet columns for each class. See https://doc.topquadrant.com/7.0/ontologies/#Modify_Menu and https://doc.topquadrant.com/7.0/importing/#Import_Property_Shapes_from_Spreadsheet_Columns. You will get a dialog where you will be able to indicate that some columns are relationships.

4. Make your classes “public” as described here: https://doc.topquadrant.com/7.0/ontologies/#SHACL_requirements_for_working_with_Reference_Datasets_and_Data_Graphs

5. Create a data graph in EDG, base it on the ontology above. 

6. Import spreadsheet data as described here: https://doc.topquadrant.com/7.0/importing/#Import_Spreadsheet_using_Pattern

Each import creates instances of a single class that you select in the wizard.

Further, you must have a row to import for each resource you want to create. 

For example, if there are two columns Person and Organization and you want this to be a relationship. You need to first create organizations, then people. You would need to import the spreadsheet twice - first time to create organizations (you would not map the Person column), second time to create people and build the relationship to organizations (you would map both, person and organization columns).

7. Switch to the SPARQL Layout to query your data

--
You received this message because you are subscribed to the Google Groups "TopBraid Suite Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to topbraid-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/topbraid-users/bed54728-0ad3-4681-9508-de86ff2fe22an%40googlegroups.com.