Transformation Engine - RDF Conversion Tool

From LinDA Wiki
Revision as of 22:07, 15 January 2016 by Admin (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

- Back to LinDA Tools

LinDA Transformation Engine - RDF Conversion Tool


Transform data into RDF in a few, simple steps. Suitable for non-experts

By using LinDA Transformation engine, users can publish their data as linked data in a few, simple steps. Simply connect to your database, select the data table you want and make your mappings to popular and standardized vocabularies. LinDA assists even more by providing automatic suggestions to the mapping through its Suggest API.



  • Transform CSV and EXCEL files into RDF
  • Transform RDB into RDF
  • Wizard-like process with explanations on each step
  • Select which columns you want to transform to RDF
  • Map data columns to properties of popular Linked data vocabularies
  • Get automatic suggestions for the most suitable linked data property for your column
  • Reconciliation of data rows to DBpedia resources using ‘Dbpedia Lookup’
  • Auto-suggest the data type of each row (e.g xsd:decimal)
  • Suggest or manually set the rdf:type of each data row based on classes of popular Linked data vocabularies
  • Ability to manually tweak the final RDF
  • Download final RDF file as a rdf dump
  • Download the transformation mappings as an R2RML file
  • Store output RDF to LinDA Data sources
  • Support for R2RML mappings
  • Saving/reload state of transformation mappings
  • View list of saved transformation mappings. Right-click and reload them or run them
  • Reconciliation of data rows to any linked data resource (not only DBPedia), including other resources created within LinDA
  • Ability to set a predefined list of column names to automatically set as auto-inference / reconciliation

Example Tutorials

Example 1 (Monthly List of employees)

  • Step 1 - Upload your file
  • Go to Transformations you will be directed to the transformation tool.
  • Choose the data format you would like to transform to RDF, select CSV/Excel
  • Click on the ‘Browse’ button and select the file you would like to upload.
  • Click on ‘Upload file’
  • If the content of the file is shown in a peculiar way in the DATA VIEW section please enter the parameters (line end, quote char, delimiter,escape) according to those used in the file and click ‘Apply’
  • Click ‘next step’
  • Step 2 column selection
    • Select the columns you would like to be added to the RDF.
    • Select all
    • Click ‘next step’
  • Step 3 specify your RDF Subject:
    • Construct the RDF subject by adding in the Base URL input field the following:
    • Select ‘Employee ID’ from the blue labelled columns and drag it into the subject URI input field
    • Click ‘next step’
  • Step 4 specify the RDF predicate
    • The tool will automatically load possible properties from adequate ontologies for you.
    • The tool will find matches for each entry except for ‘Home Address’
    • Select the following for the columns
      • Firstname : firstname - friend of a friend
      • Lastname: lastname - friend of a friend
      • Home Address : In the input field enter ‘address’ and choose address from DBpedia ontology
      • Keep the rest of the suggested selections as is.
    • Click ‘next step’
  • Step 5 Specify the RDF Object:
    • Select ‘add data type’ from the drop down menu of each column for the following columns:Employee ID, firstname, lastname, home address, salary, gender..
    • Select ‘add URI’ for the following columns: City, Country, and Year.
    • Click ‘next step’
  • Step 6 Enrich the RDF graph
    • In the input field enter a term that describes your data.
    • Enter employee
    • Select Employee from AKT Reference Ontology and has employee / employer from Agent Relationship Ontology
    • Click ‘next step’
  • Step 7 publish/save RDF
    • Enter a name for the RDF file, enter ‘Employees_05_2015’
    • Click on publish to triple store
    • Click ok on message
  • Step 8 Go to Data Sources and search for the name you entered. Search for ‘Employees_05_2015’

Example 2 (Table of restaurant contracts)

  • Step 0 - Invoke server
  • Step 1 - Select database
    • Set the connection to the database details.Please contact us for testing this example on our online site.
    • Then click Connect; the first ten data records should appear.
    • Click next step
  • Step 2 - Additional Tables
    • You can skip this in this simple example.Click next step
  • Step 3 – Table and Column Selection
    • All columns are selected by default, nothing needs to be done.Click next step
  • Step 4 – Specify the RDF Subject
    • Specify the base URI, e.g. „“.
    • This causes the remaining fields to be filled with reasonable values,
    • i.e., the Primary Key column is employed for subject URI generation.
    • Click next step.
  • Step 5 – Specify the RDF Predicate
    • For every selected column (i.e., for all columns in this example),an RDF property needs to be found by the Vocabulary Oracle.
    • The search fields for the oracle are pre-filled with the colum names; this partially needs to be corrected:
      • name: leave as is
      • start_date: replace by „start date“ (underscore → space)
      • contract_ID: replace by „contract“
      • country: leave as is
      • federal_state: replace by „federal state“ (underscore → space)
      • foreign_ID: replace by „id“.
    • For all columns except the last, stick to the suggestion found by the vocabulary server.
    • For column foreign_ID, select the „Vocabulary for Attaching Essential Metadata“ entry from the menu.
    • Click next step.
  • Step 6 – Specify the RDF Object
    • change the menu setting „no action“ on the top of every column as follows:
      • name: add data type / string
      • start_date: add data type / string
      • contract_ID: add data type / decimal
      • country: add URIs
      • federal_state: add URIs.
    • The setting „add URIs“ causes the literals in that column to be replaced by URIs (currently from dbpedia).
    • Click next step.
  • Step 7 – Enrich the RDF Graph
    • Enter „contract“ into the search box; a scroll box with suggestions for the subject class appears.
    • Double click on the first one.
    • Click next step.
  • Step 8 – Publish
    • Alter the proposed dataset name if desirable.
    • Then click any choice of
      • Publish to Triple Store
      • Download RDF
      • Download R2RML (for a script for a SPARQL server).
    • Click Restart for another data transformation.