NLL2RDF

NLL2RDF is an automated framework for RDF-based licenses specifications starting from natural language texts. It provides both human users, and automated systems with a support to generate machine readable representations of licensing terms. The CC REL and the ODRL vocabularies are adopted to specify the licenses in RDF.
NLL2RDF takes as input the natural language definition of the licensing terms to be “translated” into RDF. NLL2RDF accesses such NL text T and applies some preprocessing steps: tokenization, lemmatization, part-of-speech tagging (relying on Stanford Parser). After that, a classification step is performed using kernel methods to assign the CC REL and the ODRL relations and values to the licenses. In particular we use Support Vector Machine (libSVM 3.1219).
Finally, the RDF version of the license is generated filling a pre-defined RDF template.

Authors:

  • Elena Cabrio and Serena Villata (INRIA Sophia Antipolis, France)
  • Alessio Palmero Aprosio (Machine Linking Srl, Italy)